compute_ece

probly.evaluation.active_learning.compute_ece = <flextype.singledispatch.Flexdispatch object>

Compute Expected Calibration Error.

Parameters:
  • probs – Class probability matrix of shape (n, n_classes).

  • y_true – Ground-truth integer class labels of shape (n,).

  • n_bins – Number of confidence bins for the ECE histogram.

Returns:

ECE value in [0, 1].

Examples using probly.evaluation.active_learning.compute_ece

Active Learning with sklearn - Margin vs Random

Active Learning with sklearn - Margin vs Random

Active Learning with PyTorch - BADGE Selection

Active Learning with PyTorch - BADGE Selection