Personalized and Private Peer-to-Peer Machine Learning Modelisation with Python

This work is an implementation of the concepts from Personalized and Private Peer-to-Peer Machine Learning (Aurélien Bellet and al.). You can find the original paper there : http://proceedings.mlr.press/v84/bellet18a/bellet18a.pdf

Required packages

You need to install numpy, matplotlib and sklearn.

The data

You need to download the dataset ml-100k. It is available here : https://grouplens.org/datasets/movielens/100k/

The config.py file

You MUST state the path to your dataset folder in the variable PATH_TO_DATA (end with "/"). You can also change the default variables of the script.

Running the code

To run the code, just run the main.py file.

$ python main.py

Graphs pictures will be saved in your working folder.