Alibi
A Python library for explaining machine learning models using black-box, white-box, local, and global interpretation methods.
What is Alibi?
Alibi is a Python library dedicated to explaining machine learning models. It implements a wide range of algorithms for model inspection and interpretation, helping data scientists and ML engineers understand model predictions, debug performance, and ensure regulatory compliance. The library supports both classification and regression models across tabular, text, and image data.
Machine learning practitioners, data scientists, and researchers who need to interpret, debug, or validate their models, especially those working on projects requiring transparency, fairness, or regulatory compliance.
Developers choose Alibi for its comprehensive, production-ready collection of state-of-the-art explanation methods, its clean API inspired by scikit-learn, and its focus on high-quality implementations that work across diverse data types and model architectures.
Overview
Algorithms for explaining machine learning models
Use Cases
Best For
- Explaining black-box model predictions for regulatory compliance
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
