GraphSage

A PyTorch implementation of the paper https://www-cs.stanford.edu/people/jure/pubs/graphsage-nips17.pdf

Requirements

Pytorch >=1.1.0

DGL: 0.4.3.post2

Results

The node classification results include two parts:

• Full graph training on Cora citation dataset
• Minibatch training on Reddit dataset

Full graph training

Run with following to train a GraphSage network on the Cora dataset:

python train_full_cora.py

Notice: This version not performs neighbor sampling (i.e. Algorithm 1 in the paper) so we feed the model with the entire graph and corresponding feature matrix.

• GraphSage-Mean: ~ 80.4%
• GraphSage-GCN: ~ 83.4%
• GraphSage-Pool: ~ 72.5%

Minibatch training

Run with following to train a GraphSage network on the Reddit dataset:

python train_sampling_reddit.py

Notice: This version performs neighbor sampling in a layer-wise way (i.e. Algorithm 2 in the paper) so we feed the model with blocks (undirected bipartite graph).

• GraphSage-Mean: ~ 96.33%
• GraphSage-GCN: ~ 94.53%
• GraphSage-Pool: ~ 88.35%

To-do:

• LSTM aggregator

• Minibatch training: Inductive graph splitting

• Unsupervised training