Repository with an example of training a GCN using DGL

Introduction

This repository provides an example of training a graph neural network using the DGL framework on the Cora dataset.

The Cora dataset consists of 2708 scientific publications classified into one of seven classes. The citation network consists of 5429 links. Each publication in the dataset is described by a 0/1-valued word vector indicating the absence/presence of the corresponding word from the dictionary.

Goal: predict which class an article belongs to.

Repo structure

• dgl_cora.ipynb file with a description of the model architecture, its training, as well as an inference.

• models directory for storing a model file in pt format , as well as a python file with architecture.

  • GCN.py file describing the architecture of the GCN model.
  • APPNP.py file describing the architecture of the APPNP model.
  • CRD_CLS.py file describing the architecture of the CRD_CLS model.
  • gcn_model.pt GCN model file in pt format.
  • appnp_model.pt APPNP model file in pt format.
  • crd_cls_model.pt CRD_CLS model file in pt format.

Usage

This repository is used to provide files to the dl-benchmark repository.