fairlearn

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

What is fairlearn?

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

Overview of fairlearn

What it is

Fairlearn is a Python package that enables developers of artificial intelligence (AI) systems to assess their system's fairness and mitigate any observed unfairness issues. It contains mitigation algorithms as well as metrics for model assessment. The repository also includes Jupyter notebooks with examples of Fairlearn usage. The package follows the approach known as group fairness, which asks which groups of individuals are at risk for experiencing harms. Fairness is defined in terms of impact on people, focusing on two kinds of harms: allocation harms (when AI systems extend or withhold opportunities, resources, or information) and quality-of-service harms (whether a system works as well for one person as it does for another). The project is licensed under the MIT license.

Key features

  • Metrics for assessing which groups are negatively impacted by a model, and for comparing multiple models in terms of various fairness and accuracy metrics.
  • Algorithms for mitigating unfairness in a variety of AI tasks and along a variety of fairness definitions.
  • Jupyter notebooks with examples of Fairlearn usage are included in the repository.
  • A comprehensive user guide and API reference are available.
  • Community support via Stack Overflow (tag: fairlearn) and Discord.

Use cases

Fairlearn can be used in any scenario where an AI system may produce unfair outcomes. Common applications include hiring, school admissions, and lending, where allocation harms may occur. It is also applicable to any system where quality-of-service harms may arise, meaning the system does not work equally well for all individuals. The package is designed for data scientists and developers who need to assess and mitigate unfairness in their machine learning models.

Getting started

Install via pip: pip install fairlearn For a quick start, visit the Quickstart guide on the Fairlearn website. Example notebooks are available in the repository.

License and pricing

The project is licensed under the MIT license. It is open source and free to use.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/fairlearnfairlearn.svg)](https://yunodb.com/tools/fairlearnfairlearn?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/fairlearnfairlearn?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/fairlearnfairlearn.svg" alt="Featured on YunoDB" height="20"></a>