CEML

Counterfactuals for Explaining Machine Learning models

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 (1.5.0)
• PyTorch (2.0.1)
• Keras & Tensorflow (2.13.1)

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.

Installation

Note: Python 3.8 is required!

Tested on Ubuntu -- note that some people reported problems with some dependencies on Windows!

PyPI

pip install ceml

Note: The package hosted on PyPI uses the cpu only. If you want to use the gpu, you have to install CEML manually - see next section.

Git

Download or clone the repository:

git clone https://github.com/andreArtelt/ceml.git
cd ceml

Install all requirements (listed in requirements.txt):

pip install -r requirements.txt

Note: If you want to use a gpu/tpu, you have to install the gpu version of jax, tensorflow and pytorch manually. Do not use pip install -r requirements.txt.

Install the toolbox itself:

pip install .

Quick example

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from sklearn.tree import DecisionTreeClassifier

from ceml.sklearn import generate_counterfactual


if __name__ == "__main__":
    # Load data
    X, y = load_iris(return_X_y=True)
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=4242)

    # Whitelist of features - list of features we can change/use when computing a counterfactual
    features_whitelist = None   # We can use all features

    # Create and fit model
    model = DecisionTreeClassifier(max_depth=3)
    model.fit(X_train, y_train)

    # Select data point for explaining its prediction
    x = X_test[1,:]
    print("Prediction on x: {0}".format(model.predict([x])))

    # Compute counterfactual
    print("\nCompute counterfactual ....")
    print(generate_counterfactual(model, x, y_target=0, features_whitelist=features_whitelist))

Documentation

Documentation is available on readthedocs:https://ceml.readthedocs.io/en/latest/

License

MIT license - See LICENSE

How to cite?

You can cite CEML by using the following BibTeX entry:

@misc{ceml,
        author = {André Artelt},
        title = {CEML: Counterfactuals for Explaining Machine Learning models - A Python toolbox},
        year = {2019 - 2023},
        publisher = {GitHub},
        journal = {GitHub repository},
        howpublished = {\url{https://www.github.com/andreArtelt/ceml}}
    }

Third party components

• numpy
• scipy
• jax
• cvxpy
• scikit-learn
• sklearn-lvq
• PyTorch
• tensorflow