This is a Pytorch implementation of paper: DropEdge: Towards Deep Graph Convolutional Networks on Node Classification
The data format is this repo. The load_data() function in the utils.py,
this function helps organize data in preparation of training.
# load_data() function in train.py
adj, features, y_train, y_val, y_test, train_mask, val_mask, test_mask = load_data(FLAGS.dataset)