eval_toolkit.metric_specs#
First-party MetricSpec instances consumed by
scorecard(). The namespace ships threshold-free
specs only (Decision R). Each unparameterized metric is a module-level
singleton; parameterized metrics (currently just the ECE family) are
factory callables with LRU-cached identity.
|
Brier-score singleton spec (mean squared error of probabilistic predictions). |
|
Factory for an ECE |
|
Canonicalize a parameterized |
|
PR-AUC singleton spec (area under the precision-recall curve). |
|
ROC-AUC singleton spec (area under the receiver-operating-characteristic curve). |