Metrics

The imblearn.metrics module includes score functions, performance metrics and pairwise metrics and distance computations.

Classification metrics

See the Metrics section of the user guide for further details.

classification_report_imbalanced(y_true, …)

Build a classification report based on metrics used with imbalanced dataset

sensitivity_specificity_support(y_true, …)

Compute sensitivity, specificity, and support for each class

sensitivity_score(y_true, y_pred, *[, …])

Compute the sensitivity

specificity_score(y_true, y_pred, *[, …])

Compute the specificity

geometric_mean_score(y_true, y_pred, *[, …])

Compute the geometric mean.

macro_averaged_mean_absolute_error(y_true, …)

Compute Macro-Averaged Mean Absolute Error (MA-MAE) for imbalanced ordinal classification.

make_index_balanced_accuracy(*[, alpha, squared])

Balance any scoring function using the index balanced accuracy

Pairwise metrics

See the Pairwise metrics section of the user guide for further details.

Metrics to perform pairwise computation.

ValueDifferenceMetric(*[, n_categories, k, r])

Class implementing the Value Difference Metric.