A reimplementation of the paper Semi-Supervised Classification with Graph Convolutional Networks by Kipf and Welling 2016 in PyTorch. See also their official implementation here. I will solely focus on reproducing the results for the cora dataset, which I pushed into the git repository as it is not very large.
The early stopping class is taken from the repository of Bjarten. I only changed the process of saving the model slightly.
Run:
make environment
source activate pytorch-gcn
make requirements