autointent.metrics.decision_accuracy#

autointent.metrics.decision_accuracy(y_true, y_pred)#

Calculate decision accuracy. Supports both multiclass and multilabel.

The decision accuracy is calculated as:

\[\text{Accuracy} = \frac{\sum_{i=1}^N \mathbb{1}(y_{\text{true},i} = y_{\text{pred},i})}{N}\]

where:

  • \(N\) is the total number of samples,

  • \(y_{\text{true},i}\) is the true label for the \(i\)-th sample,

  • \(y_{\text{pred},i}\) is the predicted label for the \(i\)-th sample,

  • \(\mathbb{1}(\text{condition})\) is the indicator function that equals 1 if the condition

is true and 0 otherwise.

Parameters:
  • y_true (autointent.custom_types.ListOfGenericLabels) – True values of labels

  • y_pred (autointent.custom_types.ListOfGenericLabels) – Predicted values of labels

Returns:

Score of the decision accuracy

Return type:

float