williamleif / graphsage-simple

Simple reference implementation of GraphSAGE.
993 stars 242 forks source link

Reference PyTorch GraphSAGE Implementation

Author: William L. Hamilton

Basic reference PyTorch implementation of GraphSAGE. This reference implementation is not as fast as the TensorFlow version for large graphs, but the code is easier to read and it performs better (in terms of speed) on small-graph benchmarks. The code is also intended to be simpler, more extensible, and easier to work with than the TensorFlow version.

Currently, only supervised versions of GraphSAGE-mean and GraphSAGE-GCN are implemented.

Requirements

pytorch >0.2 is required.

Running examples

Execute python -m graphsage.model to run the Cora example. It assumes that CUDA is not being used, but modifying the run functions in model.py in the obvious way can change this. There is also a pubmed example (called via the run_pubmed function in model.py).