Marko Djordjevic 40d6d1739e fix(ml): add windowed feature flattening for inference parity ... The model was trained on 94-candle sliding windows flattened to 2820 features (94 candles x 30 features). Inference was sending raw per-candle features (27 columns). Changes: - Rewrite preprocessing to return (X, window_times) tuple - Add sliding window creation with correct feature ordering - Fill missing columns (average, barCount) with 0 for feature parity - Fill NaN from indicator warmup with 0 instead of dropping rows - Always compute all indicators (including MFI) for feature parity - Update predict and batch predict endpoints for new signature		2026-02-15 22:07:06 +01:00
..
__pycache__	fix(api): add GET /api/charts/[id] and fix batch prediction	2026-02-15 21:49:22 +01:00
__init__.py	feat(ml): add database schema, config parser, and DVC setup	2026-02-15 12:08:53 +01:00
annotation_ingestion.py	fix(ml): parse datetime column and fix TA-Lib pattern names	2026-02-15 21:13:20 +01:00
config.py	feat(ml): add database schema, config parser, and DVC setup	2026-02-15 12:08:53 +01:00
db.py	fix(ml): complete ML pipeline fixes and setup	2026-02-15 21:29:54 +01:00
main.py	fix(ml): add windowed feature flattening for inference parity	2026-02-15 22:07:06 +01:00
preprocessing.py	fix(ml): add windowed feature flattening for inference parity	2026-02-15 22:07:06 +01:00