/egnn_DL_project

Project for the Deep Learning class (2022/2023)

Primary LanguagePythonMIT LicenseMIT

E(n) Equivariant Graph Neural Networks

Using the official implementation (Pytorch 1.7.1) of:

E(n) Equivariant Graph Neural Networks
Victor Garcia Satorras, Emiel Hogeboom, Max Welling
https://arxiv.org/abs/2102.09844

Project:

Applying the EGNN model to the standard AIDS graphs benchmark dataset. The project works without node positions by setting all of them to 0. By doing so the Equivariant Graph Convolutional Layer works like a standard Graph Convolutional Layer.
To run the project just type:

cd models/egnn_clean
python3 main.py