AlphaStream

ML trading signals that show their work.

AlphaStream is a Python package for ML-based trading signal generation. It's open-source, maintained, and used by quant practitioners who want explainable signals backed by a rigorous feature engineering pipeline.

ML signal engines for trading don't require a hedge fund infrastructure team. A well-engineered Python package with the right architecture can be built, maintained, and extended by a single practitioner — and released as open source without compromising the core thesis. The ensemble architecture and SHAP explainability are what make AlphaStream a practitioner tool, not a toy.

• 200+ technical indicators computed via TA-Lib, pandas, and custom implementations
• 5 ML models per instrument/timeframe: XGBoost (primary), LightGBM (speed-optimized), Random Forest (confidence calibration), Ridge Regression (trend baseline), Ensemble Voter (learned weights)
• Backtesting engine with walk-forward validation and held-out test sets — no look-ahead bias
• SHAP value explainability: every signal maps to contributing features with importance scores
• Clean CLI and programmatic Python API with full documentation and strategy examples