Graph Neural Networks

This repository contains an introduction to Graph Neural Networks (GNNs) and a simple implementation of a GNN for node classification on the Cora dataset. The code is based on the PyTorch Geometric (PyG) library.

Requirements

• PyTorch 1.4.0
• NetworkX 2.4
• Scikit-learn 0.22.1
• numpy 1.18.1
• PYG 1.4.3

Authors

• Lakhder Amine - Github - LinkedIn

License

This project is licensed under the MIT License - see the LICENSE.md file for details