How mayongupta/quantmltoolkit works
quant-ml-toolkit is a Python toolkit for building and honestly evaluating price-data ML models, with leakage-safe walk-forward validation as its central capability. Its documented workflow generates or receives OHLCV data, derives past-only features and forward-looking targets, creates chronological folds (optionally purged with an embargo), trains fresh models per fold, and reports classification or regression metrics. The README also describes feature-importance reporting. The example’s source was not sampled, so workflow links rely on README statements; implementation links use the excerpts shown.
Main entry points: evaluation.py Classification runner · synthetic_data.py Synthetic OHLCV · features.py Feature engineering · labels.py Target labels
Components 10
Data Preparation
- Synthetic OHLCV
quant_ml/synthetic_data.py - Feature engineering
quant_ml/features.py - Target labels
quant_ml/labels.py
Leak-Free Validation
- Walk-forward splits
quant_ml/validation.py
Modeling
- Model factories
quant_ml/models.py - Feature importance
quant_ml/models.py
Evaluation
- Classification runner
quant_ml/evaluation.py - Regression runner
quant_ml/evaluation.py - Fold metrics
quant_ml/evaluation.py
Other components
- Pipeline user
Connections 14
- Pipeline user → Synthetic OHLCV: generates data
README.md - Synthetic OHLCV → Feature engineering: provides OHLCV
README.md - Synthetic OHLCV → Target labels: provides OHLCV
README.md - Feature engineering → Classification runner: supplies features
README.md - Target labels → Classification runner: supplies targets
README.md - Feature engineering → Regression runner: supplies features
README.md - Target labels → Regression runner: supplies targets
README.md - Walk-forward splits → Classification runner: supplies folds
README.md - Walk-forward splits → Regression runner: supplies folds
README.md - Model factories → Classification runner: creates models
quant_ml/evaluation.py - Model factories → Regression runner: creates models
quant_ml/evaluation.py - Classification runner → Fold metrics: reports metrics
quant_ml/evaluation.py - Regression runner → Fold metrics: reports metrics
quant_ml/evaluation.py - Pipeline user → Feature importance: requests importance
README.md