fast_delong

Note

This page is a reference documentation. It only explains the function signature, and not how to use it. Please refer to the Habitat Guide and Python API guide (v2.0) for usage.

fast_delong(predictions_sorted_transposed: ndarray, label_1_count: int) → Tuple[ndarray, ndarray][source]

Compute AUCs and the DeLong covariance for one or more classifiers.

The fast version of DeLong’s method for the covariance of unadjusted AUC (Sun & Xu, IEEE Signal Processing Letters 21(11), 2014).

Parameters:

predictions_sorted_transposed – Scores of shape (n_classifiers, n_examples), sorted so that the examples with label 1 come first.
label_1_count – Number of positive examples (m in the paper).

Returns:

Tuple (aucs, delong_covariance) with aucs of shape (n_classifiers,) and the covariance matrix of shape (n_classifiers, n_classifiers) (scalar when k == 1).