4 projects
fairmedfm
Fairness metrics for any binary classification or segmentation model: subgroup AUC, accuracy, calibration and Dice gaps, equalized odds and equity-scaled Dice
rvcbench
Comprehensive voice cloning evaluation: automatic speech metrics and ready-to-use benchmark datasets
evalrx
Failure case analysis in the era of LLM.
evalvitals
Failure case analysis in the era of LLM.