Adrian Flores Cortes
d015e2b0f3
feat(ml-engine): Phase 4 - PostgreSQL migration, dynamic OOS, data pipeline
...
- Fix database.py: Add DatabaseConnection alias for backward compat
- Fix train_symbol_timeframe_models.py: Use PostgreSQLConnection + native queries
- Fix run_oos_backtest.py: Fix broken import + add dynamic OOS support
- Update data_splitter.py: split_dynamic_oos() method (from previous session)
- Update validation_oos.yaml: Dynamic OOS config + all 6 symbols enabled
- Create ingest_ohlcv_polygon.py: Standalone Polygon→PostgreSQL ingestion script
- Fix .gitignore: /data/ instead of data/ to not ignore src/data/
- Add untracked src/ modules: backtesting, data, llm, models (attention/metamodel/strategies)
- Add aiohttp, sqlalchemy, psycopg2-binary to requirements.txt
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-27 04:39:05 -06:00
75c4d07690
feat: Initial commit - ML Engine codebase
...
Hierarchical ML Pipeline for trading predictions:
- Level 0: Attention Models (volatility/flow classification)
- Level 1: Base Models (XGBoost per symbol/timeframe)
- Level 2: Metamodels (XGBoost Stacking + Neural Gating)
Key components:
- src/pipelines/hierarchical_pipeline.py - Main prediction pipeline
- src/models/ - All ML model classes
- src/training/ - Training utilities
- src/api/ - FastAPI endpoints
- scripts/ - Training and evaluation scripts
- config/ - YAML configurations
Note: Trained models (*.joblib, *.pt) are gitignored.
Regenerate with training scripts.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-18 04:27:40 -06:00