https://github.com/ing-bank/probatus

Validation (like Recursive Feature Elimination for SHAP) of (multiclass) classifiers & regressors and data used to develop them.
https://github.com/ing-bank/probatus

binary-classifiers data-analysis data-science feature-elimination machine-learning multi-class-classification recursive-feature-elimination regressors shap statistics tree-model

Last synced: 3 days ago
JSON representation

Validation (like Recursive Feature Elimination for SHAP) of (multiclass) classifiers & regressors and data used to develop them.

• Host: GitHub
• URL: https://github.com/ing-bank/probatus
• Owner: ing-bank
• License: mit
• Created: 2020-11-09T13:25:51.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2024-08-19T12:00:08.000Z (4 months ago)
• Last Synced: 2024-12-21T12:05:10.711Z (10 days ago)
• Topics: binary-classifiers, data-analysis, data-science, feature-elimination, machine-learning, multi-class-classification, recursive-feature-elimination, regressors, shap, statistics, tree-model
• Language: Python
• Homepage: https://ing-bank.github.io/probatus
• Size: 12.4 MB
• Stars: 132
• Watchers: 9
• Forks: 40
• Open Issues: 17
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md

Awesome Lists containing this project

README

        


[![pytest](https://github.com/ing-bank/probatus/workflows/Development/badge.svg)](https://github.com/ing-bank/probatus/actions?query=workflow%3A%22Development%22)

[![PyPi Version](https://img.shields.io/pypi/pyversions/probatus)](#)

[![PyPI](https://img.shields.io/pypi/v/probatus)](#)

[![PyPI - Downloads](https://img.shields.io/pypi/dm/probatus)](#)

![GitHub contributors](https://img.shields.io/github/contributors/ing-bank/probatus)

# Probatus

## Overview

**Probatus** is a python package that helps validate regression & (multiclass) classification models and the data used to develop them. Main features:

- [probatus.interpret](https://ing-bank.github.io/probatus/api/model_interpret.html) provides shap-based model interpretation tools

- [probatus.sample_similarity](https://ing-bank.github.io/probatus/api/sample_similarity.html) to compare two datasets using resemblance modelling, f.e. `train` with out-of-time `test`.

- [probatus.feature_elimination.ShapRFECV](https://ing-bank.github.io/probatus/api/feature_elimination.html) provides cross-validated Recursive Feature Elimination using shap feature importance.

## Installation

```bash

pip install probatus

```

## Documentation

Documentation at [ing-bank.github.io/probatus/](https://ing-bank.github.io/probatus/).

You can also check out blog posts about Probatus:

-  [Open-sourcing ShapRFECV — Improved feature selection powered by SHAP.](https://medium.com/ing-blog/open-sourcing-shaprfecv-improved-feature-selection-powered-by-shap-994fe7861560)

-  [Model Explainability — How to choose the right tool?](https://medium.com/ing-blog/model-explainability-how-to-choose-the-right-tool-6c5eabd1a46a)

## Contributing

To learn more about making a contribution to Probatus, please see [`CONTRIBUTING.md`](CONTRIBUTING.md).