search

snap-stanford / GraphGym

Platform for designing and evaluating Graph Neural Networks (GNN)
Other
1.69k stars 185 forks source link

question about IDGNN #30

Closed Chen-Cai-OSU closed 3 years ago

Chen-Cai-OSU commented 3 years ago

Hello,

I am interested in reproducing the IDGNN's results on graph classification. I looked at the code and had a few quick questions 1) are all configurations listed in https://github.com/snap-stanford/GraphGym/blob/master/run/grids/IDGNN/graph_enzyme.txt? i.e., the major arguments I need to change is dataset.augment_feature? I am mainly interested in reproducing results at Table 6.

2) there are ID-GNN and ID-GNN-Fast. Are both implemented in this repository?

3) how is the heterogeneous message passing is implemented?

Thank you very much!!

JiaxuanYou commented 3 years ago

Thanks for the questions.

  1. For all the IDGNN experiments, please check https://github.com/snap-stanford/GraphGym/blob/master/run/grids/IDGNN/graph.txt graph_enzyme is the experiments for enzyme dataset only (where a smaller model is used)

  2. Yes, ID-GNN-Fast is implemented as well. dataset.augment_feature = 'node_identity' is for ID-GNN-Fast.

  3. Please checkout this file https://github.com/snap-stanford/GraphGym/blob/master/graphgym/contrib/layer/idconv.py