skactiveml.pool.expected_average_precision#

skactiveml.pool.expected_average_precision(classes, probas)[source]#

Calculate the expected average precision [1].

Parameters

classesarray-like, shape=(n_classes,): Holds the label for each class.
probasarray-like of shape (n_samples, n_classes): Class membership probabilities for each sample.

Returns

scorenp.ndarray of shape=(n_samples,): The expected average precision score of all samples in candidates.

References

1: H. Wang, X. Chang, L. Shi, Y. Yang, and Y.-D. Shen. Uncertainty Sampling for Action Recognition via Maximizing Expected Average Precision. In Int. Jt. Conf. Artif. Intell., pages 964–970, 2018.