delong_roc_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.

delong_roc_test(ground_truth: ndarray, predictions_one: ndarray, predictions_two: ndarray) → float[source]

Compute the p-value for the hypothesis that two ROC AUCs differ.

Parameters:

  • ground_truth – Binary labels (0/1), both classes present.

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

  • predictions_two – Probability-of-class-1 scores of the second model.

Returns:

Two-sided p-value of the paired DeLong test.