Variational Graph Auto-encoder in Pytorch Geometric

This respository implements variational graph auto-encoder in Pytorch Geometric, adapted from the autoencoder example code in pyG. For details of the model, refer to Thomas Klpf's original paper.

Requirements

• Python >= 3.6
• Pytorch == 1.5
• Pytorch Geometric == 1.5
• scikit-learn
• scipy

How to run

1. Configure the arguments in config/vgae.yaml file. You can also make your own config file.

2. Specify the config file and run the training script.

python train.py --load_config config/vgae.yaml

Result

We follow the arguments set as the original paper and the results is shown below.

Dataset	AUC	AP
Cora	0.903	0.911
Citeseer	0.869	0.879
Pubmed	0.948	0.948