Add electricity price ingestion and feature pipeline.

Introduce ENTSO-E data retrieval with layered caching, robust bidding-zone and missing-data handling, and persist model-ready features with detailed architecture/developer documentation.

Made-with: Cursor

electricity_price_predictor/scripts/build_feature_store.py

@@ -0,0 +1,41 @@
+import argparse
+import os
+
+import pandas as pd
+
+from electricity_price_predictor.db import get_engine
+from electricity_price_predictor.pipeline import run_feature_pipeline
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="Fetch ENTSO-E inputs and build feature store.")
+    parser.add_argument("--country-code", required=True, help="ENTSO-E bidding zone code, e.g. DE_LU")
+    parser.add_argument("--start", required=True, help="Inclusive start datetime, e.g. 2026-01-01T00:00:00Z")
+    parser.add_argument("--end", required=True, help="Exclusive end datetime, e.g. 2026-02-01T00:00:00Z")
+    parser.add_argument("--cache-ttl-hours", type=int, default=24, help="Decorator cache TTL in hours")
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+    api_key = os.getenv("ENTSOE_API_KEY")
+    if not api_key:
+        raise RuntimeError("ENTSOE_API_KEY environment variable is required.")
+
+    engine = get_engine()
+    start = pd.Timestamp(args.start, tz="UTC")
+    end = pd.Timestamp(args.end, tz="UTC")
+
+    features = run_feature_pipeline(
+        engine=engine,
+        entsoe_api_key=api_key,
+        country_code=args.country_code,
+        start=start,
+        end=end,
+        cache_ttl_hours=args.cache_ttl_hours,
+    )
+    print(f"Persisted {len(features)} feature rows for {args.country_code}.")
+
+
+if __name__ == "__main__":
+    main()

electricity_price_predictor/scripts/init_db.py

@@ -0,0 +1,23 @@
+from pathlib import Path
+
+from sqlalchemy import text
+
+from electricity_price_predictor.db import get_engine
+
+
+def main() -> None:
+    engine = get_engine()
+    schema_path = Path(__file__).resolve().parents[1] / "sql" / "001_electricity_price_schema.sql"
+    sql = schema_path.read_text(encoding="utf-8")
+
+    with engine.begin() as conn:
+        for statement in sql.split(";"):
+            stmt = statement.strip()
+            if stmt:
+                conn.execute(text(stmt))
+
+    print("Schema initialized for electricity price predictor.")
+
+
+if __name__ == "__main__":
+    main()