AffectLog Widgets

This repository contains the affectlog_widgets library, which is used for creating interactive, visual Trustworthy AI tools. This project provides a set of widgets for model assessment, fairness analysis, error detection, and more, built on top of machine learning frameworks.

Requirements for Development

To contribute to affectlog_widgets, make sure to install the following dependencies.

Testing Dependencies

For running tests with code coverage and mocking, use the following:

pytest==7.0.1
pytest-cov
pytest-mock==3.6.1

Parser Dependencies

For parsing requirements, use:

requirements-parser==0.2.0

Package Building

Ensure you have the necessary tools for building the package:

wheel

Machine Learning & Fairness Libraries

Install the following libraries to enable machine learning and fairness evaluation tools:

fairlearn==0.7.0
ml-wrappers>=0.4.0
sktime
pmdarima

Note: For Windows users running Python 3.7, add the following fix for joblib compatibility:

joblib<1.3.0; python_version <= '3.7' and sys_platform == 'win32'

Notebook Testing

To test notebooks and perform notebook validation, install:

nbformat
papermill
scrapbook
jupyter
nbval

Documentation Dependencies

For generating documentation, install the following:

docutils<=0.19
sphinx==5.0.2
sphinx-gallery==0.10.0
pydata-sphinx-theme==0.7.2

Setting up the Development Environment

Follow these steps to set up your local development environment:

1. Clone the repository:

   git clone <repository_url>
   cd affectlog_widgets
   

2. Install Dependencies: You can install all dependencies via pip:

   pip install -r requirements.txt
   

3. Run Tests: To ensure everything is working, run the tests:

   pytest
   

Local Installation

For local development, ensure that trustworthyai is available in your project. You can use the relative path for local development:

../trustworthyai/.

Add this path when running your local environment so the necessary modules can be accessed.

Contributing

If you're interested in contributing, please create a feature branch from main, make your changes, and open a pull request. Ensure that all tests pass and documentation is updated where necessary.

License

This project is licensed under the MIT License. See the LICENSE file for details.