dataeval.bias¶

Check for skewed or imbalanced datasets and incomplete feature representation.

This may impact model performance.

Classes¶

`Balance`	Calculates mutual information (MI) between factors (class label, metadata, label/image properties).
`Diversity`	Computes diversity and classwise diversity for discrete/categorical variables.
`Parity`	Calculate statistical parity using Bias-Corrected Cramér's V.

`BalanceOutput`	Output class for the `Balance` bias evaluator.
`DiversityOutput`	Output class for the `Diversity` bias evaluator.
`ParityOutput`	Output class for the `Parity` bias evaluator.