bqlearn.corruptions.make_instance_dependent_label_noise

bqlearn.corruptions.make_instance_dependent_label_noise(noise_prob, y, noise_matrix='uniform', *, random_state=None, labels=None)

Corrupt the labels given a noise transition matrix and a noise probability function.

Parameters:

noise_probarray-like of shape (n_samples, )

The noise probabilities.

yarray-like of shape (n_samples, )

The targets.

noise_matrixstr or callable, default=”uniform”

The matrix representing probabilities transition between clean labels and noisy labels. If noise_matrix is a string, it must be one of the metrics in noise_matrices.NOISE_MATRIX_FUNCTIONS. If noise_matrix is a callable, it should take a number of classes, a noise probability, and optionally a random state as the input and outputs a numpy array of shape (n_classes, n_classes).

random_stateint or RandomState, default=None

Controls the noise matrix construction.

labelsarray-like of shape (n_classes), default=None

List of labels to index the matrix. This may be used to reorder or select a subset of labels. If None is given, those that appear at least once in y and are used in sorted order.

Returns:

y_corruptndarray of shape (n_samples,)

The corrupted targets.