# fairlearn/fairlearn

A Python package to assess and improve fairness of machine learning models.

Repository: https://github.com/fairlearn/fairlearn
Canonical: https://ross.abutalabs.com/products/fairlearn
Homepage: https://fairlearn.org
Language: Python
License: MIT
License Family: permissive
Topics: fairness-ml, fairness-ai, fairness, machine-learning, artificial-intelligence, unfairness-mitigation, fairness-assessment, ai, responsible-ai, harms, group-fairness, ai-systems
Last push: 2026-08-24T17:23:49+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 55, longevity 100
- inputs: {"age_days": 3033, "days_push": 9, "days_rel": 86, "gap_med": 231, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2277, forks 513 (observed 2026-08-28T04:06:33.496833+00:00)

## What it is
Fairlearn is a Python package for assessing and mitigating fairness issues in machine learning models. It provides fairness metrics for evaluating model behavior across groups and algorithms for reducing unfairness under group fairness definitions.

## Use cases
- measure fairness metrics across demographic groups in a classifier
- mitigate bias in a loan default prediction model
- compare models by accuracy and fairness trade-offs
- detect quality-of-service disparities in an ML system
- apply group fairness constraints during model training
- audit hiring or admissions models for allocation harms

## When to choose
- you need quantitative fairness metrics and mitigation algorithms in a Python/scikit-learn workflow
- you want to assess group-level harms in classification or regression models
- you need an actively maintained, MIT-licensed fairness toolkit

## When to avoid
- you need causal fairness analysis or individual fairness definitions
- you expect fairness to be solved purely by running code without sociotechnical context
- you need fairness tooling outside the Python ecosystem

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-science, monitoring
- domain: machine-learning, artificial-intelligence, data-science
- platform: python, cross-platform
- tags: fairness, responsible-ai, group-fairness, bias-mitigation, fairness-metrics, scikit-learn, algorithms

## Member repositories
- fairlearn/fairlearn (main) score 84

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:33.496833+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:41:26.167792+00:00, confidence not recorded.
  - readme: https://github.com/fairlearn/fairlearn (fetched 2026-08-28T04:06:33.496833+00:00, sha 6ce2aa4beb09)
  - homepage: https://fairlearn.org (fetched 2026-08-29T10:21:50.344283+00:00, sha 8b841b100477)
  - registry_pypi: https://pypi.org/pypi/fairlearn/json (fetched 2026-08-29T10:21:50.353736+00:00, sha 103761675fae)
- Data as of 2026-08-30T08:39:29.467469+00:00.