pairwise_delong_test

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.

pairwise_delong_test(y_true: ndarray, scores_a: ndarray, scores_b: ndarray) → DelongResult[source]

Compare two models’ ROC AUCs on the same subjects (paired DeLong test).

Parameters:

  • y_true – Binary ground-truth labels (0/1), both classes present.

  • scores_a – Probability-of-class-1 scores of the first model.

  • scores_b – Probability-of-class-1 scores of the second model, aligned to scores_a.

Returns:

Frozen DelongResult.