conn2res.performance.balanced_accuracy_score

conn2res.performance.balanced_accuracy_score(y_true, y_pred, sample_weight=None, adjusted=None, **kwargs)[source]

Balance accuracy score. Good to deal with imbalanced datasets.

Parameters:

  • y_true (numpy.ndarray) – Ground truth target values.

  • y_pred (numpy.ndarray) – Predicted target values.

  • sample_weight (numpy.ndarray) – Sample weights.

  • adjusted (bool) – When true, the result is adjusted for chance, so that random performance would score 0, while keeping perfect performance at a score of 1.

Returns:

score – A floating point value.

Return type:

float.