skclean.simulate_noise.flip_labels_cc

skclean.simulate_noise.flip_labels_cc(y, lcm, random_state=None)

Class Conditional Noise: general version of flip_labels_uniform, a sample’s probability of getting mislabelled and it’s new (noisy) label depends on it’s true label, but not features.

Parameters
  • Y (np.ndarray) – 1-D array of labels

  • lcm (np.ndarray) – Short for Label Confusion Matrix. lcm[i,j] denotes the probability of a sample with true label i getting mislabelled as j.

  • random_state (int, default=None) – Set this value for reproducibility

Returns

Yn – 1-D array of flipped labels

Return type

np.ndarray