A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in datasets and models.
Interpret is an interpretable machine learning library and glassbox model framework. It provides toolkits for training inherently transparent models and applying post-hoc explanation techniques to make machine learning predictions human-understandable. The framework distinguishes itself by integrating differential privacy into the training of interpretable models to prevent sensitive data from leaking through explanations. It also features a visualization tool for rendering interactive decision paths and model behavior. The library covers model explainability through feature importance calcu
Algorithms for explaining machine learning models
Source code/webpage/demos for the What-If Tool
The main features of pair-code/what-if-tool are: Explainability and Fairness, Explainable AI Libraries, Guardrails and AI Safety.
Open-source alternatives to pair-code/what-if-tool include: trusted-ai/aif360 — A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and… seldonio/alibi — Algorithms for explaining machine learning models. interpretml/interpret — Interpret is an interpretable machine learning library and glassbox model framework. It provides toolkits for training… pytorch/captum — Captum is an open-source library for explaining model predictions by attributing them to input features, neurons, and… marcotcr/lime — This project is an agnostic model interpretability framework and explainability tool designed to provide local… fairlearn/fairlearn — A Python package to assess and improve fairness of machine learning models.
