ag_sklearn: Extending Diffprivlib for Enhanced Scikit-Learn Functionality

Welcome to ag_sklearn, an open-source repository that extends the capabilities of the widely recognized diffprivlib, a general-purpose library focused on differential privacy. This repository is an initiative by the open-source community at CodePeak 2023, aiming to broaden the scope of diffprivlib by integrating more functionalities from Scikit-Learn (sklearn).

The core of ag_sklearn is based on diffprivlib v0.6. We credit the original authors and contributors of diffprivlib for their exceptional work, and our project builds upon this foundation to explore new frontiers in privacy-preserving machine learning.

Background: Diffprivlib v0.6

• Core Functionality: diffprivlib is designed for experimenting with, investigating, and developing differential privacy applications. It supports a range of machine learning tasks like classification, clustering, and more.
• Python Compatibility: Compatible with Python versions 3.8 to 3.11.

ag_sklearn: What's New?

ag_sklearn extends the functionality of diffprivlib by incorporating additional Scikit-Learn features, enhancing the library's capabilities in privacy-preserving machine learning. We strive to maintain compatibility with the existing diffprivlib interface while adding new functionalities and improvements.

Contributing to ag_sklearn

As an open-source project, ag_sklearn thrives on community contributions. We welcome contributions from developers, researchers, and enthusiasts in the field. Whether it's adding new features, improving documentation, or fixing bugs, your contributions are valuable to us.

To get started head over the the open issues and try to implement a solution to them under a new branch. Once complete raise a PR.

Citing

Please cite the original source of diffprivlib as can be found here.

Acknowledgements

We acknowledge the foundational work done by the creators of diffprivlib and the support of the open-source community at CodePeak 2023 in developing ag_sklearn. This collaboration symbolizes the synergy of community-driven development in advancing the field of privacy-preserving machine learning.