Welcome to CEML’s documentation!

Counterfactuals for Explaining Machine Learning models - CEML

CEML is a Python toolbox for computing counterfactuals. Counterfactuals can be used to explain the predictions of machine learing models.

It supports many common machine learning frameworks:

  • scikit-learn (0.23.1)

  • PyTorch (1.5.0)

  • Keras & Tensorflow (2.2.0)

Furthermore, CEML is easy to use and can be extended very easily. See the following user guide for more information on how to use and extend ceml.

Indices and tables