PGM-Explainer

This is the source code for the paper: PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks by Minh N. Vu and My T. Thai, proceeding in NeurIPS 2020.

Repo description:

• The folder PGM_Node contains code for the experiments on node classification tasks.
• The folder PGM_Graph contains code for the experiments on graph classification tasks.

Required libraries:

• numpy 1.19.2
• networkx 2.5
• scipy 1.5.2
• matplotlib 3.3.2
• pandas 1.1.2
• tensorboardX 2.1
• sklearn 0.23.2
• pgmpy 0.1.11
• dgl 0.4.2
• pytorch 1.6.8

To install above libraries:

pip install -r requirements.txt

For more details, please check the README in two subfolders