cleanlab
Model fairness tool
A tool for evaluating and improving the fairness of machine learning models
Official cleanlab repo is at https://github.com/cleanlab/cleanlab
57 stars
2 watching
8 forks
last commit: over 3 years agoLinked from 1 awesome list
subeeshvasu/awesome-learning-with-label-noiseRelated projects:
| Repository | Description | Stars |
|---|---|---|
| cgnorthcutt/rankpruning | An algorithm and package for handling noisy labels in binary classification problems | 82 |
 fairlearn/fairlearn | A tool to assess and mitigate unfairness in AI systems, helping developers ensure their models do not disproportionately harm certain groups of people. | 1,974 |
 koaning/scikit-fairness | A Python library providing tools and algorithms for fairness in machine learning model development | 29 |
 tensorflow/fairness-indicators | An evaluation toolkit to assess fairness in machine learning models | 343 |
 cleanlab/cleanvision | Automatically detects issues in image datasets to improve computer vision models | 1,038 |
 ucsb-nlp-chang/fairness-reprogramming | A method to improve machine learning model fairness without retraining the entire network | 15 |
 mbilalzafar/fair-classification | Provides a Python implementation of fairness mechanisms in classification models to mitigate disparate impact and mistreatment. | 190 |
 davified/clean-code-ml | Adapting clean code principles to machine learning and data science in Python | 714 |
 adebayoj/fairml | An auditing toolbox to assess the fairness of black-box predictive models | 361 |
 linkedin/lift | A tool for measuring fairness and mitigating bias in machine learning workflows | 167 |
 mcandre/linters | A community wiki for improving code quality through static analysis and style checking tools | 342 |
 x-plug/cvalues | Evaluates and aligns the values of Chinese large language models with safety and responsibility standards | 481 |
 dssg/aequitas | Toolkit to audit and mitigate biases in machine learning models | 701 |
 megantosh/fairness_measures_code | This repository contains implementations of measures to quantify discrimination in data | 38 |
 neuraxio/kata-clean-machine-learning-from-dirty-code | Converting dirty machine learning code into clean, modular, and reusable components using the Pipe and Filter Design Pattern for Machine Learning. | 18 |