Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
Job manager crashed while running this job (missing heartbeats).
Error code:   JobManagerCrashedError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

version
string
method
string
anchor_ids
list
anchor_labels
dict
background_ids
list
instance_ids
list
n_aug_anchor
int64
aug_seed
int64
eval_seed
int64
n_instances
int64
seed
int64
PARAM-BENCH-EVAL v2.0
random anchors (uniform, no type stratification, seed=42)
[ 3279, 13435, 14593, 18290, 29257, 32099, 36049, 83811, 97501, 98220 ]
{ "3279": "bspline", "13435": "plane", "14593": "plane", "18290": "bspline", "29257": "bspline", "32099": "revolution", "36049": "cylinder", "83811": "bspline", "97501": "bspline", "98220": "bspline" }
[ 47351, 64445, 77828, 70927, 20113, 61023, 52456, 94187, 49598, 61627, 83941, 88515, 87712, 91810, 4837, 18288, 38198, 17783, 91119, 85622, 57824, 87738, 10528, 54484, 102127, 76134, 73393, 79459, 36417, 81274, 77498, 3521, 32146, 21459, 10045, 89...
[ 80863, 65914, 66028, 85981, 45028, 47938, 51111, 33737, 9806, 93785, 28701, 89262, 2445, 730, 11756, 92042, 22790, 15279, 80576, 87123, 69849, 92542, 8656, 68646, 12991, 9314, 102252, 48554, 24527, 62809, 51508, 22796, 70073, 59182, 52988, 14051,...
50
20,260,731
42
300
42

PARAM-BENCH — B-Rep NURBS Surface Benchmark

Parameterization-Agnostic B-Rep Representation Learning benchmark data. Built from the ABC Dataset (STEP format).

Important: This dataset is the canonical, frozen benchmark used for the PARAM-BENCH evaluation protocol. It is not reproducible from ABC by re-running the build pipeline (sampling is stochastic, OCC version matters). Use this uploaded artifact directly.

Contents

File Size Description
surfaces.db 99 MB 100,000 NURBS surfaces (SQLite, table surfaces)
pca_augments.db 3.8 MB Pre-generated augmentations (10 anchors × 50 variants)
eval_protocol_canonical.json 58 KB Frozen eval protocol (10 anchors / 5000 background / 300 instances)
random_anchors.json 0.5 KB 10 uniform-random anchors (seed=42, no type stratification)
manifest.json 0.4 KB Build manifest
surface_vae.pt 29 MB NeuroNURBS checkpoint (parameter MSE loss)
surface_geom_vae.pt 29 MB GeomVAE checkpoint (geometric L2 loss)

Note: farthest_anchors.json (latent-FPS anchors) is deprecated and intentionally excluded.

Statistics

  • 100,000 faces from 12 ABC chunks, all converted to NURBS (BRepBuilderAPI_NurbsConvert)
  • All surfaces clamped (periodic surfaces unperiodized; original periodicity recorded in is_periodic_u/v columns)
  • Dynamic stratified sampling: type × area × trim complexity
  • Anchor faces: 3279, 13435, 14593, 18290, 29257, 32099, 36049, 83811, 97501, 98220

Quick Start

# Download the whole repo (~990 MB)
pip install huggingface_hub
huggingface-cli download Axiao1999/georep-benchmark --repo-type dataset \
    --local-dir data/benchmark

# Or individual files
curl -L https://huggingface.co/datasets/Axiao1999/georep-benchmark/resolve/main/surfaces.db -o data/benchmark/surfaces.db

Checkpoints

Two VAE models (identical architecture, 7.37M params, fdim=32/edim=256, 100K faces / 200 epochs / 70-20-10 split), differing only in loss:

  • surface_vae.ptNeuroNURBS: masked parameter MSE on control points/knots
  • surface_geom_vae.ptGeomVAE: geometric L2 on UV-sampled points (differentiable NURBS)

Expected location after download:

benchmark_neuronurbs/outputs/surface_vae_200ep/surface_vae.pt
benchmark_nurbs_geom/outputs/surface_geom_vae_200ep/surface_geom_vae.pt

Eval Protocol

eval_protocol_canonical.json (PARAM-BENCH-EVAL v2.0) freezes:

  • 10 uniform-random anchors (seed=42)
  • 5000 background faces (real class distribution)
  • 300 instances for invariance metrics
  • Pre-generated augmentation database (50 variants/anchor)

License

ABC dataset: CC-BY. Derived data distributed under CC-BY-4.0.

Downloads last month
33