← ClaudeAtlas

fairlearn-fairnesslisted

Use when a model's decisions affect people and an auditor or rule (ECOA, GDPR Art. 22, EU AI Act) needs per-group evidence, or someone reports unfair treatment of a group. Computes Fairlearn MetricFrame group metrics, mitigates with ExponentiatedGradient and adds group checks to the gate.
testland/qa-skills · ★ 0 · AI & Automation · score 72
Install: claude install-skill testland/qa-skills
# fairlearn-fairness Fairlearn provides *"Metrics - Tools to assess which groups are negatively impacted and compare models across fairness and accuracy dimensions"* and *"Algorithms - Techniques to mitigate unfairness"* per the [Fairlearn quickstart]. Two primitives: `MetricFrame` (group disaggregation) + Reductions (`ExponentiatedGradient`, `ThresholdOptimizer`). ## When to use - Pre-deployment: assert per-group accuracy / selection rate disparities are within budget. - Bias incident triage: a stakeholder reports the model is unfair to group X; produce evidence + a mitigated comparison. - Compliance evidence (ECOA, GDPR Art. 22, EU AI Act high-risk systems): group-disaggregated metrics + mitigation provenance. ## Step 1 - Install ```bash pip install fairlearn # OR conda install -c conda-forge fairlearn ``` Per the [Fairlearn quickstart]. ## Step 2 - Compute disaggregated accuracy ```python from fairlearn.metrics import MetricFrame from sklearn.metrics import accuracy_score from sklearn.tree import DecisionTreeClassifier classifier = DecisionTreeClassifier(min_samples_leaf=10, max_depth=4) classifier.fit(X, y_true) y_pred = classifier.predict(X) mf = MetricFrame( metrics=accuracy_score, y_true=y_true, y_pred=y_pred, sensitive_features=sex, ) print(mf.by_group) print(f"Disparity (max-min): {mf.difference()}") ``` Per the [Fairlearn quickstart]. `sensitive_features` can be a Series or a 2-D array for intersectional analysis (sex × race). ## St