gin

Graph Isomorphism Network (GIN)

• Paper link: arXiv OpenReview
• Author's code repo: https://github.com/weihua916/powerful-gnns.

Dependencies

• scikit-learn

Install as follows:

pip install scikit-learn

How to run

Run with the following for bioinformatics graph classification (available datasets: MUTAG (default), PTC, NCI1, and PROTEINS)

python3 train.py --dataset MUTAG

NOTE: Users may observe results fluctuate due to the randomness with relatively small dataset. In consistence with the original paper, five social network datasets, 'COLLAB', 'IMDBBINARY' 'IMDBMULTI' 'REDDITBINARY' and 'REDDITMULTI5K', are also available as the input. Users are encouraged to update the script slightly for social network applications, for example, replacing sum readout on bioinformatics datasets with mean readout on social network datasets and using one-hot encodings of node degrees by setting "degree_as_nlabel=True" in GINDataset.

Summary (10-fold cross-validation)

Dataset	Result
MUTAG	~89.4
PTC	~68.5
NCI1	~82.9
PROTEINS	~74.1