The Classification and Regression Performance Metrics library

For classification/ranking: ROC AUC, BEDROC AUC, Enrichment Factor, Robust Initial Enhancement, Power Metric, Matthews' Correlation Coefficient, Platt scaling.

For regression: Root Mean Squared Error, Mean Absolute Error, r^2 coefficient of determination, Raw Regression Error Characteristic Curve.

Also features a TopKeeper module; to keep in memory the top 'k' scored items when dealing with very large datasets.