responsible-ai-widgets/responsibleai
Ilya Matiach ab28cd3fca
remove python 3.6 from builds, tests and packages (#2442)
2023-12-06 11:34:24 -05:00
..
responsibleai Backend and OD Notebook Dataset fixes (#2360) 2023-09-29 14:35:00 -04:00
tests fix build error due to changed exception messages and UCI repository (#2413) 2023-11-13 20:39:08 -05:00
.gitignore Add License and Requirements to Source Distribtion (#1522) 2022-07-05 13:08:46 -04:00
LICENSE Add License and Requirements to Source Distribtion (#1522) 2022-07-05 13:08:46 -04:00
MANIFEST.in Add License and Requirements to Source Distribtion (#1522) 2022-07-05 13:08:46 -04:00
README.md remove python 3.6 from builds, tests and packages (#2442) 2023-12-06 11:34:24 -05:00
requirements-dev.txt fix build failures due to new joblib 1.3.0 release failing on windows with python 3.7 (#2155) 2023-06-29 10:53:45 -04:00
requirements.txt Update requirements.txt to upgrade dice-ml to 0.11.0 (#2397) 2023-10-30 20:35:38 +00:00
setup.py remove python 3.6 from builds, tests and packages (#2442) 2023-12-06 11:34:24 -05:00

README.md

Responsible AI Model Analysis SDK for Python

This package has been tested with Python 3.7, 3.8, 3.9 and 3.10

The Responsible AI Model Analysis SDK enables users to analyze their machine learning models in one API. Users will be able to analyze errors, explain the most important features, compute counterfactuals and run causal analysis using a single API.

Highlights of the package include:

  • explainer.add() explains the model
  • counterfactuals.add() computes counterfactuals
  • error_analysis.add() runs error analysis
  • causal.add() runs causal analysis

Supported scenarios, models and datasets

responsibleai supports computation of Responsible AI insights for scikit-learn models that are trained on pandas.DataFrame. The responsibleai accept both models and pipelines as input as long as the model or pipeline implements a predict or predict_proba function that conforms to the scikit-learn convention. If not compatible, you can wrap your model's prediction function into a wrapper class that transforms the output into the format that is supported (predict or predict_proba of scikit-learn), and pass that wrapper class to modules in responsibleai.

Currently, we support datasets having numerical and categorical features. The following table provides the scenarios supported for each of the four responsible AI insights:-

RAI insight Binary classification Multi-class classification Multilabel classification Regression Timeseries forecasting Categorical features Text features Image Features Recommender Systems Reinforcement Learning
Explainability Yes Yes No Yes No Yes No No No No
Error Analysis Yes Yes No Yes No Yes No No No No
Causal Analysis Yes No No Yes No Yes (max 5 features due to expensiveness) No No No No
Counterfactual Yes Yes No Yes No Yes No No No No

The source code can be found here: https://github.com/microsoft/responsible-ai-toolbox/tree/main/responsibleai