v0.12.0

This release focuses on improving scikit-learn compatibility and fixing several core components. Key highlights include support for scikit-learn 1.6 and various improvements to metrics and documentation.

scikit-learn compatibility

• Fairlearn is now compatible with scikit-learn 1.6: #1451 by Tamara Atanasoska.

• The fairlearn.preprocessing.CorrelationRemover estimator is now fully scikit-learn compatible: #1452 by Tamara Atanasoska.

Other improvements

• Fixed MetricFrame bootstrapping with missing sensitive feature values: #1450 by Tahar Allouche.

• Added weighted objective’s curve to ThresholdOptimizer’s plot: #1445 by Tahar Allouche.

• Corrected CorrelationRemover’s mathematical formulation in documentation: #1438 by Tahar Allouche.

• Fixed CorrelationRemover’s default constructor and added input validation: #1439 by Tahar Allouche.

• Updating the documentation infrastructure for ease of use: #1421 by Tamara Atanasoska.