Upload netherlands-electricity-demand-forecaster model

README.md

@@ -0,0 +1,148 @@
+---
+language:
+- en
+- nl
+license: apache-2.0
+tags:
+- time-series-forecasting
+- energy-forecasting
+- electricity-demand
+- day-ahead-prices
+- lightgbm
+- tabular-regression
+- conformal-prediction
+- netherlands
+- europe
+- entsoe
+pipeline_tag: tabular-regression
+datasets:
+- hsilvosa/entsoe-day-ahead
+metrics:
+- mae
+- rmse
+- pinball_loss
+- winkler_score
+model-index:
+- name: netherlands-demand-forecaster
+  results:
+  - task:
+      type: tabular-regression
+      name: Electricity Demand Forecasting
+    dataset:
+      name: ENTSO-E Netherlands Bidding Zone (NL)
+      type: hsilvosa/entsoe-day-ahead
+    metrics:
+    - name: MAE
+      type: mae
+      value: 228.837
+    - name: RMSE
+      type: rmse
+      value: 760.680
+    - name: Empirical Interval Coverage (80% Nominal)
+      type: coverage
+      value: 72.4%
+---
+# Electricity Demand Forecaster for Netherlands (NL)
+
+High-accuracy calibrated quantile LightGBM model for forecasting Netherlands **demand**.
+Resolution: 15-minute intervals. Trained on multi-year data (2023–2026) from **ENTSO-E**,
+featuring multi-scale lags, cyclical encodings, and **conformal calibration**
+for well-calibrated 80% prediction intervals ($P10, P50, P90$).
+
+## Model Highlights
+
+- **Country / Zone**: Netherlands (`NL`)
+- **Target**: Electricity Demand in `MW`
+- **Resolution**: 15-minute intervals
+- **Outputs**: Point forecast ($P50$), 80% prediction interval ($P10$ to $P90$)
+- **Algorithm**: LightGBM Multi-Quantile Regressor with Conformal Calibration & Monotonicity
+- **Dataset**: ENTSO-E European Transparency Platform (Zone: `NL`)
+- **Training Samples**: 99,380 observations (2023–2026)
+
+## Performance & Benchmark Comparison
+
+Evaluated on out-of-sample test sets against official seasonal persistence benchmarks:
+
+| Metric | LightGBM Forecaster | 7-Day Seasonal Persistence | Improvement |
+|---|---:|---:|---:|
+| **MAE** | **228.837 MW** | 1655.395 MW | **+86.2%** |
+| **RMSE** | **760.680 MW** | — | — |
+- **WAPE**: 1.68%
+| **P10 Pinball Loss** | 147.679 | — | — |
+| **P90 Pinball Loss** | 54.656 | — | — |
+| **P10–P90 Interval Coverage** | **72.4%** | — | Target: 75–85% |
+| **Winkler Score** | 2023.352 | — | — |
+
+## Quickstart: Python Inference
+
+```python
+import pandas as pd
+from huggingface_hub import hf_hub_download
+import joblib
+
+# 1. Download model artifacts
+model_path = hf_hub_download(repo_id="ORGANIZATION/netherlands-demand-forecaster", filename="models.joblib")
+models = joblib.load(model_path)
+
+# 2. Predict P10, P50 (point), and P90 quantiles
+X_test = pd.read_csv("sample_input.csv")
+p10 = models[0.1].predict(X_test)
+p50 = models[0.5].predict(X_test)
+p90 = models[0.9].predict(X_test)
+
+print("Forecast Point Estimate:", p50[:5])
+print("80% Lower Bound (P10):", p10[:5])
+print("80% Upper Bound (P90):", p90[:5])
+```
+
+## Features Used
+
+The model uses 37 leakage-safe features:
+- `hour`
+- `quarter`
+- `day_of_week`
+- `day_of_year`
+- `month`
+- `is_weekend`
+- `is_holiday`
+- `is_morning_peak`
+- `is_evening_peak`
+- `sin_hour`
+- `cos_hour`
+- `sin_day_of_week`
+- `cos_day_of_week`
+- `sin_day_of_year`
+- `cos_day_of_year`
+- `lag_1h`
+- `lag_2h`
+- `lag_3h`
+- `lag_4h`
+- `lag_24h`
+- `lag_48h`
+- `lag_7d`
+- `lag_14d`
+- `diff_1h`
+- `diff_2h`
+- `diff_24h`
+- `diff_7d`
+- `acceleration_1h`
+- `ema_4step`
+- `ema_12step`
+- `rolling_std_4step`
+- `rolling_mean_24h`
+- `rolling_std_24h`
+- `rolling_min_24h`
+- `rolling_max_24h`
+- `rolling_mean_7d`
+- `rolling_std_7d`
+
+## Intended Use & Advisory
+
+This model is intended for research, energy market analytics, grid load planning,
+and educational forecasting demonstrations. It is advisory only and not intended
+for automated trading execution or real-time grid dispatch.
+
+## Citation & Attribution
+
+Data published under the ENTSO-E Transparency framework:
+- Transparency Platform: https://transparency.entsoe.eu/

config.json

@@ -0,0 +1,68 @@
+{
+  "target": "demand",
+  "country_code": "NL",
+  "zone_key": "NL",
+  "model_name": "lightgbm-quantile",
+  "model_version": "20260820180933",
+  "feature_names": [
+    "hour",
+    "quarter",
+    "day_of_week",
+    "day_of_year",
+    "month",
+    "is_weekend",
+    "is_holiday",
+    "is_morning_peak",
+    "is_evening_peak",
+    "sin_hour",
+    "cos_hour",
+    "sin_day_of_week",
+    "cos_day_of_week",
+    "sin_day_of_year",
+    "cos_day_of_year",
+    "lag_1h",
+    "lag_2h",
+    "lag_3h",
+    "lag_4h",
+    "lag_24h",
+    "lag_48h",
+    "lag_7d",
+    "lag_14d",
+    "diff_1h",
+    "diff_2h",
+    "diff_24h",
+    "diff_7d",
+    "acceleration_1h",
+    "ema_4step",
+    "ema_12step",
+    "rolling_std_4step",
+    "rolling_mean_24h",
+    "rolling_std_24h",
+    "rolling_min_24h",
+    "rolling_max_24h",
+    "rolling_mean_7d",
+    "rolling_std_7d"
+  ],
+  "n_estimators": 180,
+  "learning_rate": 0.04,
+  "num_leaves": 31,
+  "training_rows": 99380,
+  "start_year": 2023,
+  "end_year": 2026,
+  "metrics": {
+    "mae": 228.8369113435931,
+    "rmse": 760.6798309337345,
+    "wape": 0.016802358270333835,
+    "pinball_p10": 147.6788275273391,
+    "pinball_p90": 54.65633858613622,
+    "interval_coverage": 0.7244405087747545,
+    "winkler_score": 2023.3516611347532,
+    "mape": 3.8506640888814605
+  },
+  "baseline_mae": 1655.3946271936886,
+  "calibrator": {
+    "is_fitted": true,
+    "q_correction": -3.0787313982800697,
+    "target_coverage": 0.8
+  }
+}

inference.py

@@ -0,0 +1,29 @@
+# Standalone inference helper for Demand Forecaster
+import json
+from pathlib import Path
+import numpy as np
+import pandas as pd
+import joblib
+
+def load_forecaster(model_dir="."):
+    path = Path(model_dir)
+    config = json.loads((path / "config.json").read_text())
+    models = joblib.load(path / "models.joblib")
+    features = config["feature_names"]
+    q_correction = config["calibrator"]["q_correction"] if config.get("calibrator") else 0.0
+
+    def predict(df_features, apply_calibration=True):
+        X = df_features[features]
+        p10 = models[0.1].predict(X)
+        p50 = models[0.5].predict(X)
+        p90 = models[0.9].predict(X)
+        stacked = np.sort(np.vstack([p10, p50, p90]), axis=0)
+        p10, p50, p90 = stacked[0], stacked[1], stacked[2]
+        if apply_calibration:
+            p10 -= q_correction
+            p90 += q_correction
+            stacked_cal = np.sort(np.vstack([p10, p50, p90]), axis=0)
+            p10, p50, p90 = stacked_cal[0], stacked_cal[1], stacked_cal[2]
+        return pd.DataFrame({"p10": p10, "p50_point": p50, "p90": p90}, index=df_features.index)
+
+    return predict

models.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7dd2f178def0de075749b88d6712246bce8cbc6ad4e31a7d96ae95384064af5
+size 1581827

sample_input.csv

@@ -0,0 +1,25 @@
+hour,quarter,day_of_week,day_of_year,month,is_weekend,is_holiday,is_morning_peak,is_evening_peak,sin_hour,cos_hour,sin_day_of_week,cos_day_of_week,sin_day_of_year,cos_day_of_year,lag_1h,lag_2h,lag_3h,lag_4h,lag_24h,lag_48h,lag_7d,lag_14d,diff_1h,diff_2h,diff_24h,diff_7d,acceleration_1h,ema_4step,ema_12step,rolling_std_4step,rolling_mean_24h,rolling_std_24h,rolling_min_24h,rolling_max_24h,rolling_mean_7d,rolling_std_7d
+6,0,5,319,11,1,0,0,0,1.0,6.123233995736766e-17,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,11722.386,11760.103,11766.858,12031.712,12534.19,12813.422,11572.908,11685.951,-37.716999999998734,-44.47199999999975,-279.23199999999997,-113.04299999999967,-30.961999999997715,11779.960257581632,11837.636902434691,40.02005549988108,15023.71878125,2108.8178766825954,11603.092,17810.047,14146.746081845236,2022.355527629804
+6,1,5,319,11,1,0,0,0,1.0,6.123233995736766e-17,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,11753.771,11759.492,11667.679,12007.967,12842.115,13081.165,11660.844,11814.113,-5.720999999999549,86.09200000000055,-239.0500000000011,-153.26900000000023,-97.53399999999965,11878.31655454898,11866.592917444737,123.86637709007077,15018.423583333331,2115.761615867911,11603.092,17810.047,14147.420104166667,2021.5717534450744
+6,2,5,319,11,1,0,0,0,1.0,6.123233995736766e-17,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,11780.176,11741.801,11619.456,11922.091,13248.989,13491.556,11749.362,11973.525,38.375,160.71999999999935,-242.56700000000092,-224.16300000000047,-83.96999999999935,11973.919132729388,11905.166776299395,162.7431228327848,15010.873666666666,2124.882854647522,11603.092,17810.047,14148.099388392855,2020.8115238464281
+6,3,5,319,11,1,0,0,0,1.0,6.123233995736766e-17,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,11816.721,11732.803,11603.092,11808.647,13675.916,13971.684,11956.304,12168.421,83.91799999999967,213.628999999999,-295.76800000000003,-212.1170000000002,-45.792999999999665,12088.702279637635,11959.891426099488,186.09665275974325,15000.580833333333,2135.8719468898216,11603.092,17810.047,14148.860571428571,2020.0028150417877
+7,0,5,319,11,1,0,0,0,0.9659258262890683,-0.25881904510252063,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,12025.851,11722.386,11760.103,11766.858,14313.444,14565.719,12236.173,12598.587,303.46500000000015,265.7480000000014,-252.27499999999964,-362.41399999999885,341.1819999999989,12255.369767782582,12043.81136054572,209.0673924677883,14988.38765625,2146.8268260823834,11603.092,17810.047,14149.677635416665,2019.2255292432164
+7,1,5,319,11,1,0,0,0,0.9659258262890683,-0.25881904510252063,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,12117.323,11753.771,11759.492,11667.679,14715.548,15057.648,12414.302,12855.244,363.5519999999997,357.83100000000013,-342.09999999999854,-440.9420000000009,369.27299999999923,12405.13946066955,12133.962535846376,231.8716740852777,14970.849635416665,2159.23988141094,11603.092,17810.047,14150.26338095238,2018.726654435938
+7,2,5,319,11,1,0,0,0,0.9659258262890683,-0.25881904510252063,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,12260.877,11780.176,11741.801,11619.456,15176.069,15513.793,12532.536,13006.243,480.70100000000093,519.0760000000009,-337.72400000000016,-473.70700000000033,442.32600000000093,12577.809676401732,12242.093684177704,240.79305625302274,14951.279499999999,2170.0649033734717,11603.092,17810.047,14150.892120535715,2018.250917861866
+7,3,5,319,11,1,0,0,0,0.9659258262890683,-0.25881904510252063,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,12505.371,11816.721,11732.803,11603.092,15553.792,15849.117,12680.909,13164.458,688.6499999999996,772.5679999999993,-295.3250000000007,-483.5490000000009,604.732,12786.00780584104,12373.818501996517,259.4295357257584,14929.636124999999,2178.1456971566454,11603.092,17810.047,14151.734038690476,2017.692739411205
+8,0,5,319,11,1,0,1,0,0.8660254037844387,-0.4999999999999998,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,12629.794,12025.851,11722.386,11760.103,15830.352,16061.323,12937.122,13453.022,603.9429999999993,907.4079999999994,-230.97099999999955,-515.9000000000015,300.47799999999916,13003.515083504624,12520.888886304745,305.13677853083936,14906.469291666666,2183.258628914359,11603.092,17810.047,14152.699614583335,2017.1430047324218
+8,1,5,319,11,1,0,1,0,0.8660254037844387,-0.4999999999999998,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,12836.815,12117.323,11753.771,11759.492,16051.919,16162.27,13020.734,13575.569,719.4920000000002,1083.0439999999999,-110.35100000000057,-554.8349999999991,355.9400000000005,13244.495850102778,12687.8239807194,327.9435120676824,14883.298614583335,2185.153106317325,11603.092,17810.047,14153.69491964286,2016.7072814570454
+8,2,5,319,11,1,0,1,0,0.8660254037844387,-0.4999999999999998,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,13098.305,12260.877,11780.176,11741.801,16299.309,16217.741,13102.488,13782.166,837.4279999999999,1318.1290000000008,81.5679999999993,-679.6779999999999,356.72699999999895,13478.62111006167,12863.513983685645,319.22005767047443,14860.151635416667,2184.4128946109145,11603.092,17810.047,14154.898900297618,2016.2713552269636
+8,3,5,319,11,1,0,1,0,0.8660254037844387,-0.4999999999999998,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,13329.776,12505.371,11816.721,11732.803,16545.848,16274.292,13128.878,14016.533,824.4050000000007,1513.0550000000003,271.5560000000023,-887.6549999999988,135.75500000000102,13688.016266037004,13038.682447734007,290.8417593211279,14836.222468749998,2181.061460815093,11603.092,17810.047,14156.23762202381,2015.8701719238834
+9,0,5,319,11,1,0,1,0,0.7071067811865476,-0.7071067811865475,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,13605.967,12629.794,12025.851,11722.386,16614.978,16218.071,13264.131,14123.681,976.1730000000007,1580.116,396.90699999999924,-859.5500000000011,372.2300000000014,13896.361759622203,13218.712840390313,256.03941268395835,14811.879052083334,2174.811821106313,11603.092,17810.047,14157.844767857143,2015.4803685309412
+9,1,5,319,11,1,0,1,0,0.7071067811865476,-0.7071067811865475,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,13829.809,12836.815,12117.323,11753.771,16685.004,16154.791,13347.962,14182.686,992.9939999999988,1712.485999999999,530.2130000000016,-834.7240000000002,273.5019999999986,14099.855055773322,13401.233172637958,249.6645055017418,14788.859437500001,2167.207881682092,11603.092,17810.047,14159.542630952381,2015.2069339254765
+9,2,5,319,11,1,0,1,0,0.7071067811865476,-0.7071067811865475,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14002.109,13098.305,12260.877,11780.176,16711.658,16046.293,13325.295,14352.287,903.8040000000001,1741.232,665.3649999999998,-1026.9920000000002,66.3760000000002,14278.334633463994,13577.35945377058,236.8340132232019,14766.578708333334,2158.486270620182,11603.092,17810.047,14161.325505952382,2015.0178233389886
+9,3,5,319,11,1,0,1,0,0.7071067811865476,-0.7071067811865475,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14208.88,13329.776,12505.371,11816.721,16746.187,16014.398,13266.49,14688.133,879.1039999999994,1703.509,731.7890000000025,-1421.643,54.698999999998705,14467.265180078399,13757.867383959721,228.5575436178639,14746.151656250002,2149.1438826169006,11603.092,17810.047,14163.446586309525,2014.8866675130853
+10,0,5,319,11,1,0,1,0,0.49999999999999994,-0.8660254037844387,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14405.095,13605.967,12629.794,12025.851,16822.601,15993.658,13210.181,14830.498,799.1279999999988,1775.3009999999995,828.9429999999993,-1620.316999999999,-177.0450000000019,14642.98470804704,13934.589940273609,221.1510961504145,14726.988916666669,2139.3018640012197,11603.092,17810.047,14165.887172619046,2014.7918751710563
+10,1,5,319,11,1,0,1,0,0.49999999999999994,-0.8660254037844387,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14546.054,13829.809,12836.815,12117.323,16748.229,15854.917,13184.159,14790.303,716.2450000000008,1709.2389999999996,893.3119999999999,-1606.1440000000002,-276.748999999998,14706.392824828225,14067.961487923823,151.38663227028493,14705.935833333335,2128.378667698078,11603.092,17810.047,14168.255214285715,2014.6020877848587
+10,2,5,319,11,1,0,1,0,0.49999999999999994,-0.8660254037844387,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14750.661,14002.109,13098.305,12260.877,16792.497,15798.979,13118.603,14661.333,748.5519999999997,1652.3559999999998,993.518,-1542.7300000000014,-155.2520000000004,14758.366094896935,14186.17141285862,65.45521504634277,14686.020177083332,2117.9870283377977,11603.092,17810.047,14170.713796130953,2014.407452864917
+10,3,5,319,11,1,0,1,0,0.49999999999999994,-0.8660254037844387,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14906.564,14208.88,13329.776,12505.371,16720.368,15694.39,13053.702,14739.418,697.6840000000011,1576.7880000000005,1025.9779999999992,-1685.7160000000003,-181.41999999999825,14799.944456938161,14290.193041649603,44.26649890150312,14665.914083333335,2106.9123971508607,11603.092,17810.047,14173.308601190478,2014.1732271787573
+11,0,5,319,11,1,0,1,0,0.258819045102521,-0.9659258262890682,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14801.505,14405.095,13605.967,12629.794,16715.243,15873.021,12994.259,14904.417,396.40999999999985,1195.5379999999986,842.2219999999979,-1910.1579999999994,-402.71799999999894,14870.723074162897,14395.838881395819,75.95633665295215,14647.752864583335,2096.505482840856,11603.092,17810.047,14176.170489583334,2013.946328132357
+11,1,5,319,11,1,0,1,0,0.258819045102521,-0.9659258262890682,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14836.326,14546.054,13829.809,12836.815,16713.729,15891.176,12876.269,14906.544,290.271999999999,1006.5169999999998,822.5529999999999,-2030.2749999999996,-425.9730000000018,14956.473044497741,14501.878745796463,114.30660642738745,14630.7721875,2086.159710548122,11603.092,17810.047,14179.281857142856,2013.7327276138205
+11,2,5,319,11,1,0,1,0,0.258819045102521,-0.9659258262890682,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14862.312,14750.661,14002.109,13098.305,16734.42,15789.969,12847.146,14938.824,111.65099999999984,860.2029999999995,944.4509999999991,-2091.678,-636.9009999999998,15015.546226698645,14594.5367849047,111.62392657000981,14614.005802083333,2075.684588481779,11603.092,17810.047,14182.597165178571,2013.4182317570653
+11,3,5,319,11,1,0,1,0,0.258819045102521,-0.9659258262890682,-0.9749279121818236,-0.2225209339563146,-0.7142921172691032,0.699847677146407,14976.891,14906.564,14208.88,13329.776,16751.623,15682.493,12825.573,15026.932,70.32699999999932,768.0110000000004,1069.1299999999992,-2201.3590000000004,-627.3570000000018,15082.113336019189,14684.910202611669,84.53814946043025,14597.834385416667,2065.010853196392,11603.092,17810.047,14186.07159672619,2013.124795949101

sample_prediction.json

@@ -0,0 +1,80 @@
+{
+  "p10": [
+    11807.75,
+    11968.31,
+    12102.12,
+    12173.23,
+    12424.66,
+    12579.86,
+    12788.28,
+    13043.22,
+    13132.61,
+    13364.21,
+    13508.99,
+    13652.56,
+    13827.35,
+    13969.63,
+    14078.87,
+    14235.41,
+    14325.58,
+    14340.53,
+    14394.76,
+    14431.07,
+    14468.88,
+    14540.02,
+    14555.0,
+    14647.47
+  ],
+  "p50_point": [
+    11958.18,
+    12045.23,
+    12200.73,
+    12372.99,
+    12775.0,
+    12854.91,
+    12999.33,
+    13287.36,
+    13659.98,
+    13818.07,
+    14027.31,
+    14167.33,
+    14378.99,
+    14461.85,
+    14534.06,
+    14749.41,
+    14821.17,
+    14764.82,
+    14817.77,
+    14867.88,
+    14854.67,
+    14928.91,
+    14890.53,
+    15027.52
+  ],
+  "p90": [
+    12001.23,
+    12113.94,
+    12256.82,
+    12403.55,
+    12866.88,
+    12871.65,
+    13060.02,
+    13304.75,
+    13665.4,
+    13919.79,
+    14073.66,
+    14238.12,
+    14504.35,
+    14633.76,
+    14766.65,
+    14869.79,
+    15053.02,
+    14952.09,
+    14980.28,
+    15029.77,
+    15095.32,
+    15173.47,
+    15164.58,
+    15264.01
+  ]
+}