holisticai.bias.metrics.classification_bias_metrics

holisticai.bias.metrics.classification_bias_metrics(group_a=None, group_b=None, y_pred=None, y_true=None, y_score=None, X=None, metric_type='group', **kargs)

Classification bias metrics batch computation

This function computes all the relevant classification bias metrics, and displays them as a pandas dataframe. It also includes a fair reference value for comparison.

Parameters

group_aarray-like

Group membership vector (binary)

group_barray-like

Group membership vector (binary)

y_predarray-like

Predictions vector (binary)

y_truearray-like, optional

Target vector (binary)

y_scorearray-like, optional

Probability estimates (regression)

Xarray-like, optional

Feature matrix

metric_typestr, optional

Specifies which metrics we compute ‘group’, ‘individual’, ‘equal_outcome’ , ‘equal_opportunity’

**kargsdict

Additional keyword arguments for individual metrics

Returns

pandas DataFrame

Metrics | Values | Reference