Show HN: A lightweight model monitor for scikit-learn and Keras

Canary ML, a new open-source Python library, provides lightweight drift and anomaly detection for production machine learning models built with scikit-learn, Keras, and TensorFlow. The tool wraps existing models to monitor predictions, detect data drift, and serve a dashboard, aiming to help developers maintain model performance in production.

Drift and anomaly detection for production ML models. Cutting-edge performance in a wrapper, for free. pip install canary-ml Keras/TensorFlow: pip install canary-ml keras pip install canary-ml keras python from canary ml import ModelMonitor monitor = ModelMonitor model=your model, reference data=X train, alert threshold=0.2, log path="./canary logs" drop-in replacement — monitoring is a side effect predictions = monitor.predict X new report = monitor.get report print report.psi score, report.drift detected, report.anomaly rate 0.41 True 0.032 monitor.serve dashboard port=8501