GNNsImplementation

This repository will update the implementation of variants of GNNs. Welcome to contrubute to this project!!!

Environment

The run environment is as follows:

cuda 11.3
python 3.7.13
Pytorch 1.10.1
DGL 0.8.2
PyG 2.0.4
matplotlib 3.3.4

GNN Variants

Semi-GCN: a two-layer GCN Model

Related Paper: Semi-Supervised Classification with Graph Convolutional Networks

Blog on CSDN: transport

---

GAT: a two-layer GAT Model

Related Paper: GRAPH ATTENTION NETWORKS

Blog on CSDN: transport

---

H2GCN：based on Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs

Blog on CSDN: transport

---

to be continue...

References

These project also use some others' code:

• DGL official document: transport
• PyG official document: transport