counterfactual-xai 0.0.3

Remove uncertainty from your machine learning models

Counterfactuals: Take the uncertainty out of your machine learning models

Documentation | Examples | Showcase

---

Counterfactuals is a Python library for machine learning that enables you to better understand your models. We combine several techniques to provide a comprehensive understanding of your model's predictions. With those insights you are able to eliminate uncertainties and make better decisions.

Features

CLUE: A Method for Explaining Uncertainty Estimates

CLUE is a method for explaining uncertainty estimates of machine learning models. It is based on the idea of counterfactuals and provides a comprehensive understanding of the model's predictions.

Model Paper: CLUE: A Method for Explaining Uncertainty Estimates

Installation

The current recommended way to install tinygrad is from source.

From source

git clone https://github.com/lsch0lz/counterfactuals.git
cd counterfactuals
python3 -m pip install -e .

Direct (master)

python3 -m pip install git+https://github.com/lsch0lz/counterfactuals.git

Documentation

Documentation along with a quick start guide can be found in the docs/ directory.