ranyus / FedFS

Federated Feature Selection as in the paper https://arxiv.org/abs/2109.11323
MIT License
5 stars 2 forks source link
feature-selection federated-learning

FedFS

Federated Feature Selection as in the paper https://arxiv.org/abs/2109.11323

If used in scientific publication, please cite the method and library as follows:

@misc{cgv2021federated, title={Federated Feature Selection for Cyber-Physical Systems of Systems}, author={Pietro Cassarà and Alberto Gotta and Lorenzo Valerio}, year={2021}, eprint={2109.11323}, archivePrefix={arXiv}, primaryClass={cs.LG} }

Requirements

Python>=3.6

Use the "requirements.txt" file to install all the necessary libraries.

Important: This library exploits methods from pyMIToolbox (https://github.com/tud-zih-energy/pymit). Please, be sure to install it in the python environment before running the code.