This is the Tensorflow implementation of our ICLR2018 paper: "FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling".
Instructions of the sample codes:
[For Reddit dataset]
train_batch_multiRank_inductive_reddit_Mixlayers_sampleA.py is the final model. (precomputated the AH in the bottom layer) The original Reddit data should be transferred into the .npz format using this function: transferRedditDataFormat.
train_batch_multiRank_inductive_reddit_Mixlayers_uniform.py is the codes for uniform sampling.
train_batch_multiRank_inductive_reddit_Mixlayers_appr2layers.py is the codes for 2-layer approximation.
pubmed-original**.py means the codes are used for original Cora or Pubmed datasets. Users could also change their datasets by changing the data load function from load_data() to load_data_original().
create_Graph_forGraphSAGE.py is used to transfer the data into the GraphSAGE format, so that users can compare our method with GraphSAGE. We also include the transferred original Cora dataset in this repository (./data/cora_graphSAGE).
[For pubmed or cora]
train.py is the original GCN model.
pubmed_Mix_sampleA.py The dataset could be defined in the codes, for example: flags.DEFINE_string('dataset', 'pubmed', 'Dataset string.')
pubmed_Mix_uniform.py and pubmed_inductive_appr2layers.py are similar to the ones for reddit.