nueramic_mathml.ml.metrics.roc_curve#

roc_curve(y_true, y_prob, n_thresholds=None)[source]#

Return dict with points at TPR - FPR coordinates

Parameters:

  • y_true (Tensor) – array with true values of binary classification

  • y_prob (Tensor) – array of probabilities of confidence of belonging to the 1st class

  • n_thresholds (Optional[int]) – if len(y_true) is too large, you can limit the number of threshold values

Returns:

dict with values of TPR and FPR

Return type:

Dict