Graph Convolution Network

PyTorch implementation of "Semi-Supervised Classification with Graph Convolutional Networks"

References

• https://github.com/tkipf/pygcn
• https://github.com/hazdzz/GCN

To run the node classification in CORA dataset

python main.py cora-node-classification

Results

Train-Val Loss Comparison

Train-Val Accuracy Comparison

Test Accuracy: 82.4 %

Setup

• Setup env using poetry
• Update neptune_project_name and neptune_api_token values in config/settings.yaml file