XAISuite 1.0.1

XAISuite: Training and Explanation Generation Utilities for Machine Learning Models

XAISuite: Training and Explaining Machine Learning Models

:warning: This is version 1.0A, our first public release. If there are any issues with the library, please let us know right away. Version 1.0B will be published with any edits once our research paper is published. This is the pre-research paper version.

Table of Contents

1. Introduction
2. Installation
3. Getting Started
4. How to Contribute
5. Technical Report and Citing XAISuite

Introduction

XAISuite (Explanatory Artificial Intelligence Suite) is a library for training and explaining machine learning models for tabular datasets in Python. It provides a unified interface for training any sklearn model using just a line of code and allows users to easily compare the results of different explainers. It is based on the XAISuite framework, which we propose in our paper (still in pre-publication).

XAISuite accomplishes machine learning model training and explanation generation in three steps: (1) data loading, (2) model training and explanation generation, and (3) explanation comparison. Each of these steps are delved into more detail in our documentation and in the demo tutorials. A more detailed flowchart is presented in our paper.

XAISuite was created as a helper library to a paper which compared SHAP and LIME explanations for different supervised models on tabular datasets and studied implications on model accuracy using the XAISuite framework. That paper will be published soon.

Installation

You can install the XAI Suite through PyPI:

pip install XAISuite

This will automatically install the latest version and is the reccomended way to download the library. The version on Github may not be stable. If yu already have XAISuite and want to upgrade it, do:

pip install XAISuite --upgrade

Getting Started

For example code and an introduction to the library, see the Demo Folder. The Demo folder is never fully complete and we will add more and more tutorials as the project progresses.

If you are looking for a model or dataset to use, sklearn has several cool options.

Examples of graphs and tables generated by the XAISuite Library can be found here.

How to Contribute

We welcome the contribution from the open-source community to improve the library!

To add a new functionality into the library or point out a flaw, please create a new issue on Github. We'll try to look into your requests as soon as we can. Keep in mind that, as this is an open-source project, you release any copyright protection over code you may contribute to the XAISuite Project.

Technical Report and Citing XAISuite

A paper proposing and using XAISuite to compare explanatory methods is still in pre-publication. Use the following BibTex to cite XAISuite for now:

@article{mitra2022-xaisuite,
  author    = {Shreyan Mitra and Leilani Gilpin},
  title     = {Comparison of SHAP and LIME Explanations for Supervised
Machine Learning Models Trained on Tabular Datasets},
  year      = {2022},
  doi       = {},
  url       = {},
  archivePrefix = {},
  eprint    = {},
}

Contact Us

If you have any questions, comments or suggestions, please do not hesitate to contact us at [e-mail]

License

This work is licensed under a BSD 3-Clause License.