Better Recommender systems with LightGCN - Machine Learning with Graphs project

To start, install the necessary packages:

$ pip install numpy pandas networkx cmfrec torch torch-geometric torch-sparse torch-scatter plotly jupyter 

Link to the Google Colab: https://colab.research.google.com/drive/1YD5zz2QaAP-jbOzY4mjbG6_X9EcA93lW?usp=sharing

Link to the Medium post: https://medium.com/@jn2279/better-recommender-systems-with-lightgcn-2d89d19f90b2

The repository as structured as follows:

• Folder src contains source .py scripts for LightGCN model
• Folder notebooks contains jupyter notebooks for data preprocessing, exploratory data analysis and MF models
• Folder results contains .csv files generated by executing the evaluation pipeline in the src folder
• Folder data contains the data used for the project