Upperbound Average Precision

How-To Guides

There are currently no how to’s for UAP. If there are scenarios that you want us to explain, contact us!

class dataeval.metrics.UAP

FR Test Statistic based estimate of the empirical mean precision

evaluate(labels: ArrayLike, scores: ArrayLike) → Dict[str, float]

Estimates the upperbound average precision

Parameters:

labels (ArrayLike) – A numpy array of n_samples of class labels with M unique classes.
scores (ArrayLike) – A 2D array of class probabilities per image

Returns:

uap : The empirical mean precision estimate

Return type:

Dict[str, float]

Raises:

ValueError – If unique classes M < 2