Upload spanish-day-ahead-price-forecaster model

README.md

@@ -0,0 +1,148 @@
+---
+language:
+- en
+- es
+license: apache-2.0
+tags:
+- time-series-forecasting
+- energy-forecasting
+- electricity-demand
+- day-ahead-prices
+- lightgbm
+- tabular-regression
+- conformal-prediction
+- spain
+- europe
+- entsoe
+pipeline_tag: tabular-regression
+datasets:
+- hsilvosa/entsoe-day-ahead
+metrics:
+- mae
+- rmse
+- pinball_loss
+- winkler_score
+model-index:
+- name: spain-price-forecaster
+  results:
+  - task:
+      type: tabular-regression
+      name: Electricity Price Forecasting
+    dataset:
+      name: ENTSO-E Spain Bidding Zone (ES)
+      type: hsilvosa/entsoe-day-ahead
+    metrics:
+    - name: MAE
+      type: mae
+      value: 6.039
+    - name: RMSE
+      type: rmse
+      value: 10.302
+    - name: Empirical Interval Coverage (80% Nominal)
+      type: coverage
+      value: 80.2%
+---
+# Day-Ahead Electricity Price Forecaster for Spain (ES)
+
+High-accuracy calibrated quantile LightGBM model for forecasting Spain **price**.
+Resolution: hourly 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**: Spain (`ES`)
+- **Target**: Day-Ahead Electricity Price in `EUR/MWh`
+- **Resolution**: hourly 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: `ES`)
+- **Training Samples**: 19,455 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** | **6.039 EUR/MWh** | 46.815 EUR/MWh | **+87.1%** |
+| **RMSE** | **10.302 EUR/MWh** | — | — |
+- **WAPE**: 9.79%
+| **P10 Pinball Loss** | 1.495 | — | — |
+| **P90 Pinball Loss** | 1.949 | — | — |
+| **P10–P90 Interval Coverage** | **80.2%** | — | Target: 75–85% |
+| **Winkler Score** | 34.444 | — | — |
+
+## 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/spain-price-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": "price",
+  "country_code": "ES",
+  "zone_key": "ES",
+  "model_name": "lightgbm-quantile",
+  "model_version": "20260820180911",
+  "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": 19455,
+  "start_year": 2023,
+  "end_year": 2026,
+  "metrics": {
+    "mae": 6.039063989009408,
+    "rmse": 10.301720946822583,
+    "wape": 0.09793200693108155,
+    "pinball_p10": 1.4953737829297093,
+    "pinball_p90": 1.9490205636172404,
+    "interval_coverage": 0.801562821303722,
+    "winkler_score": 34.443943465469495,
+    "mape": 268.556845469544
+  },
+  "baseline_mae": 46.81490606507185,
+  "calibrator": {
+    "is_fitted": true,
+    "q_correction": 0.3127808963314038,
+    "target_coverage": 0.8
+  }
+}

inference.py

@@ -0,0 +1,29 @@
+# Standalone inference helper for Price 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:c2cfe3d4417da7f636858e1935229a8fc2c132b56dc0266494b96de8ea8b66cf
+size 1562247

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
+0,0,4,339,12,0,0,0,0,0.0,1.0,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,65.35,80.225,85.4375,88.45666666666666,61.075,85.0475,91.6975,73.175,-14.875,-20.087500000000006,-23.972499999999997,18.522500000000008,-9.662499999999994,75.74701316546992,73.89013934524071,0.0,64.9061111111111,16.361065116030414,35.7,88.96,81.60193948412699,22.811851469196082
+1,0,4,339,12,0,0,0,0,0.25881904510252074,0.9659258262890683,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,73.8875,65.35,80.225,85.4375,55.945,79.975,87.075,64.36,8.537500000000009,-6.3374999999999915,-24.029999999999994,22.715000000000003,23.41250000000001,75.00320789928196,73.88973329212675,0.0,65.43996527777777,16.439467775858283,35.7,88.96,81.49592757936509,22.806036984012398
+2,0,4,339,12,0,0,0,0,0.49999999999999994,0.8660254037844387,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,60.07333333333333,73.8875,65.35,80.225,47.8,73.11,90.7075,62.3475,-13.814166666666672,-5.276666666666664,-25.310000000000002,28.36,-22.35166666666668,69.0312580729025,71.76413329846622,0.0,65.61197916666667,16.357190360763365,35.7,88.96,81.33520337301587,22.861562098008815
+3,0,4,339,12,0,0,0,0,0.7071067811865476,0.7071067811865476,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,56.510000000000005,60.07333333333333,73.8875,65.35,43.04,68.15,88.53000000000002,58.63333333333333,-3.5633333333333255,-17.377499999999998,-25.110000000000007,29.896666666666682,10.250833333333347,64.02275484374151,69.41734356024064,0.0,65.97489583333333,16.038330415939804,35.7,88.96,81.13164682539681,22.929756084540095
+4,0,4,339,12,0,0,0,0,0.8660254037844386,0.5000000000000001,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,55.34,56.510000000000005,60.07333333333333,73.8875,35.7,67.11,89.98,58.2775,-1.1700000000000017,-4.733333333333327,-31.409999999999997,31.7025,2.3933333333333238,60.54965290624491,67.2515983971267,0.0,66.48739583333334,15.459670943556187,35.7,88.96,80.9340873015873,23.008476265468488
+5,0,4,339,12,0,0,0,0,0.9659258262890683,0.25881904510252074,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,54.93666666666667,55.34,56.510000000000005,60.07333333333333,46.4975,72.23,92.92666666666666,52.9775,-0.403333333333336,-1.5733333333333377,-25.7325,39.94916666666666,0.7666666666666657,58.30445841041362,65.35699351551746,0.0,67.28892361111112,14.245013879932838,46.4975,88.96,80.72549603174603,23.08469880763011
+6,0,4,339,12,0,0,0,0,1.0,6.123233995736766e-17,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,59.4375,54.93666666666667,55.34,56.510000000000005,61.5625,77.9,97.47749999999999,59.5,4.500833333333333,4.097499999999997,-16.337500000000006,37.97749999999999,4.904166666666669,58.75767504624819,64.44630220543785,0.0,67.82809027777778,13.65658689676705,47.25,88.96,80.52615575396824,23.123268829832853
+7,0,4,339,12,0,0,0,0,0.9659258262890683,-0.25881904510252063,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,71.0825,59.4375,54.93666666666667,55.34,77.215,93.07,128.6575,71.935,11.644999999999996,16.14583333333333,-15.85499999999999,56.7225,7.1441666666666634,63.687605027748916,65.46725571229356,0.0,68.22475694444445,13.604845212404728,47.25,88.96,80.36904265873015,23.097058463206658
+8,0,4,339,12,0,0,1,0,0.8660254037844387,-0.4999999999999998,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,79.88499999999999,71.0825,59.4375,54.93666666666667,84.46000000000001,99.84,127.73,73.8575,8.802499999999995,20.44749999999999,-15.379999999999995,53.8725,-2.842500000000001,70.16656301664936,67.68537021809456,0.0,68.33600694444445,13.692193055274691,47.25,88.96,80.07873015873015,22.79096315165879
+9,0,4,339,12,0,0,1,0,0.7071067811865476,-0.7071067811865475,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,88.005,79.88499999999999,71.0825,59.4375,77.9275,83.8775,101.9725,27.8275,8.120000000000005,16.9225,-5.950000000000003,74.145,-0.6824999999999903,77.30193780998962,70.81146710761847,0.0,68.48371527777778,13.891370444443151,47.25,88.96,79.84227182539682,22.497806301486097
+10,0,4,339,12,0,0,1,0,0.49999999999999994,-0.8660254037844387,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,84.85249999999999,88.005,79.88499999999999,71.0825,65.975,67.38,88.0375,0.3333333333333333,-3.1525000000000034,4.967500000000001,-1.4050000000000011,87.70416666666667,-11.272500000000008,80.32216268599377,72.97162601413869,0.0,68.77225694444444,14.165278841820836,47.25,88.96,79.74036706349206,22.43565328992052
+11,0,4,339,12,0,0,1,0,0.258819045102521,-0.9659258262890682,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,82.0325,84.85249999999999,88.005,79.88499999999999,57.9925,38.1375,77.28666666666668,0.425,-2.819999999999993,-5.972499999999997,19.854999999999997,76.86166666666668,0.33250000000001023,81.00629761159627,74.36560662734811,0.0,69.44131944444443,14.404609461098717,47.25,88.96,79.70462301587301,22.42713721249495
+12,0,4,339,12,0,0,0,0,1.2246467991473532e-16,-1.0,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,74.78,82.0325,84.85249999999999,88.005,51.343333333333334,20.843333333333334,77.3475,0.7266666666666666,-7.252499999999998,-10.072499999999991,30.5,76.62083333333334,-4.4325000000000045,78.51577856695776,74.42935945390994,0.0,70.14079861111111,14.231039762508464,47.25,88.96,79.68970238095237,22.429589197293716
+13,0,4,339,12,0,0,0,0,-0.2588190451025208,-0.9659258262890683,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,65.2375,74.78,82.0325,84.85249999999999,47.25,15.442499999999999,71.75,0.86,-9.542500000000004,-16.795,31.8075,70.89,-2.2900000000000063,73.20446714017467,73.01522723023149,0.0,70.71972222222222,13.706030579827276,47.25,88.96,79.61761904761904,22.456604600809076
+14,0,4,339,12,0,0,0,0,-0.4999999999999997,-0.8660254037844388,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,55.417500000000004,65.2375,74.78,82.0325,48.44333333333333,15.175,69.0475,1.935,-9.819999999999993,-19.362499999999997,33.26833333333333,67.1125,-0.2774999999999892,66.08968028410482,70.30788457942664,0.0,71.06003472222221,13.189623637830124,48.44333333333333,88.96,79.52040178571428,22.526113468384974
+15,0,4,339,12,0,0,0,0,-0.7071067811865471,-0.7071067811865479,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,55.28,55.417500000000004,65.2375,74.78,53.2975,31.1125,74.725,14.8325,-0.13750000000000284,-9.957499999999996,22.185,59.8925,9.68249999999999,61.7658081704629,67.99590233643792,0.0,71.34489583333334,12.746301702220023,53.2975,88.96,79.43845238095238,22.589395763842877
+16,0,4,339,12,0,0,0,0,-0.8660254037844385,-0.5000000000000004,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,67.875,55.28,55.417500000000004,65.2375,69.17333333333333,64.4725,87.035,41.4625,12.594999999999999,12.457499999999996,4.700833333333335,45.5725,12.732500000000002,64.20948490227775,67.97730197698594,0.0,71.95229166666667,12.183820581091366,54.93666666666667,88.96,79.39767857142857,22.604131790705722
+17,0,4,339,12,0,0,0,0,-0.9659258262890683,-0.25881904510252063,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,78.285,67.875,55.28,55.417500000000004,79.1075,88.2875,105.3925,88.3,10.409999999999997,23.004999999999995,-9.179999999999993,17.0925,-2.1850000000000023,69.83969094136665,69.56310167283425,0.0,72.33194444444443,12.235314972150327,54.93666666666667,88.96,79.34559523809523,22.59650824938418
+18,0,4,339,12,0,0,0,0,-1.0,-1.8369701987210297e-16,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,84.3175,78.285,67.875,55.28,85.51249999999999,102.435,120.015,113.375,6.032499999999999,16.442499999999995,-16.922500000000014,6.640000000000001,-4.377499999999998,75.63081456482,71.83300910778283,0.0,72.54902777777778,12.405787274108482,54.93666666666667,88.96,79.2201488095238,22.509372593662334
+19,0,4,339,12,0,0,0,1,-0.9659258262890684,0.2588190451025203,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,87.155,84.3175,78.285,67.875,88.96,99.2175,127.58666666666666,143.34333333333333,2.8375000000000057,8.870000000000005,-10.257500000000007,-15.756666666666675,-3.194999999999993,80.24048873889201,74.19023847581624,0.0,72.61746527777778,12.484690164159131,54.93666666666667,88.96,79.02455357142856,22.294505973393328
+20,0,4,339,12,0,0,0,1,-0.8660254037844386,0.5000000000000001,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,83.0,87.155,84.3175,78.285,88.45666666666666,96.38,143.065,135.34,-4.155000000000001,-1.3174999999999955,-7.923333333333332,7.724999999999994,-6.992500000000007,81.34429324333522,75.54558640261375,0.0,72.36913194444445,12.201551495841787,54.93666666666667,88.45666666666666,78.75915674603173,21.976063000871875
+21,0,4,339,12,0,0,0,1,-0.7071067811865477,0.7071067811865474,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,79.185,83.0,87.155,84.3175,85.4375,87.24666666666667,123.49,104.3075,-3.8149999999999977,-7.969999999999999,-1.8091666666666697,19.18249999999999,0.3400000000000034,80.48057594600114,76.10549618682703,0.0,71.9828125,11.810563093793336,54.93666666666667,88.005,78.37891865079365,21.40189382106558
+22,0,4,339,12,0,0,0,1,-0.5000000000000004,0.8660254037844384,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,71.8225,79.185,83.0,87.155,80.225,85.475,102.5875,98.80666666666667,-7.362499999999997,-11.177499999999995,-5.25,3.7808333333333337,-3.5474999999999994,77.01734556760069,75.44657369654594,0.0,71.41552083333333,11.457918511098084,54.93666666666667,88.005,78.07137400793651,21.119129418723233
+23,0,4,339,12,0,0,0,0,-0.25881904510252157,0.9659258262890681,-0.433883739117558,-0.9009688679024191,-0.4363729641045924,0.8997658785476209,67.58333333333333,71.8225,79.185,83.0,65.35,76.045,99.035,92.265,-4.2391666666666765,-11.601666666666674,-10.695000000000007,6.769999999999996,3.1233333333333206,73.24374067389374,74.23684440989784,0.0,70.88878472222223,11.325134680971683,54.93666666666667,88.005,77.86301587301587,21.048363034939396

sample_prediction.json

@@ -0,0 +1,80 @@
+{
+  "p10": [
+    44.67,
+    64.02,
+    45.35,
+    46.94,
+    48.99,
+    54.94,
+    61.89,
+    73.96,
+    70.15,
+    71.77,
+    63.69,
+    59.13,
+    62.84,
+    54.19,
+    44.67,
+    49.17,
+    71.49,
+    80.08,
+    84.9,
+    86.76,
+    78.68,
+    70.38,
+    60.83,
+    52.77
+  ],
+  "p50_point": [
+    58.95,
+    71.4,
+    54.74,
+    53.33,
+    54.89,
+    58.39,
+    69.6,
+    84.3,
+    84.89,
+    83.81,
+    76.82,
+    76.44,
+    72.16,
+    62.72,
+    52.49,
+    55.9,
+    78.92,
+    87.36,
+    92.47,
+    94.15,
+    82.04,
+    75.09,
+    65.8,
+    61.49
+  ],
+  "p90": [
+    68.82,
+    80.06,
+    62.75,
+    59.48,
+    59.79,
+    66.05,
+    73.05,
+    94.3,
+    93.92,
+    97.82,
+    84.98,
+    82.7,
+    76.38,
+    67.97,
+    60.25,
+    62.74,
+    91.08,
+    99.83,
+    106.83,
+    107.18,
+    87.55,
+    83.54,
+    71.27,
+    69.49
+  ]
+}