GCN-LRP is a repository of codes and experiment results for explanation for graph convolutional networks.
See the following publications for examples of this code in use:
- GCN-LRP explanation: exploring latent attention of graph convolutional networks. Jinlong Hu, Tenghui Li, Shoubin Dong, accepted by the International Joint Conference on Neural Networks (IJCNN) 2020, 19 - 24th July, 2020, Glasgow, Scotland, UK.
gcn_lrp.py is the Python code of GCN_LRP method.
lrp_summary_test.py is the Python code of a GCN_LRP example.