underconfidence

pycalib.scoring.underconfidence(y, p_pred)

Computes the underconfidence of a classifier.

Computes the empirical underconfidence of a classifier on a test sample by evaluating the average uncertainty on the correct predictions.

Parameters

• y (array-like) – Ground truth labels

• p_pred (array-like) – Array of confidence estimates

Returns

Underconfidence

Return type

float