This is the implementation accompanying our paper, Graph Clustering with Graph Neural Networks.
First, make sure you have the requirements numpy
, scipy
, tensorflow 2
. You can always install them with pip
: pip3 install -r graph_embedding/dmon/requirements.txt
.
Then, to reproduce the paper results on the Cora graph, run
# From google-research/
python3 -m graph_embedding.dmon.train --graph_path=graph_embedding/dmon/data/cora.npz --dropout_rate=0.5
If you find DMoN useful in your research, we ask that you cite the following paper:
Tsitsulin, A., Palowitch, J., Perozzi, B., Mueller, E., (2020). Graph Clustering with Graph Neural Networks.
@inproceedings{tsitsulin2020clustering,
author={Tsitsulin, Anton and Palowitch, John and Perozzi, Bryan and M\"uller, Emmanuel}
title={Graph Clustering with Graph Neural Networks},
year = {2020},
}
For questions or comments about the implementation, please contact [email protected] or [email protected].