langfair.metrics.classification.classification.ClassificationMetrics#

class langfair.metrics.classification.classification.ClassificationMetrics(metric_type='all')#

Bases: object

__init__(metric_type='all')#

Class for pairwise classification fairness metrics.

Parameters:

metric_type (str, one of 'assistive', 'punitive', 'representation', or 'all', default='all') – A list containing name or class object of metrics.

Methods

__init__([metric_type])

Class for pairwise classification fairness metrics.

evaluate(groups, y_pred[, y_true, ratio])

Returns values of classification fairness metrics
evaluate(groups, y_pred, y_true=None, ratio=False)#

Returns values of classification fairness metrics

Parameters:

  • groups (Array-like) – Group indicators. Must contain exactly two unique values.

  • y_pred (Array-like) – Binary model predictions. Positive and negative predictions must be 1 and 0, respectively.

  • y_true (Array-like, default=None) – Binary labels (ground truth values). Positive and negative labels must be 1 and 0, respectively.

  • ratio (bool, default=False) – Indicates whether to compute the metric as a difference or a ratio

Returns:

Dictionary containing specified metric values

Return type:

dict