Skip to content

CopulaGNN: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks (ICLR 2021)

License

Notifications You must be signed in to change notification settings

jiaqima/CopulaGNN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

53 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CopulaGNN

This repo provides a PyTorch implementation for the CopulaGNN models as described in the following paper:

CopulaGNN: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks

Jiaqi Ma, Bo Chang, Xuefei Zhang, and Qiaozhu Mei. ICLR 2021.

Requirements

Most dependency packages are included in environment.yml. Run conda torch_env create -f environment.yml to install the required packages.

In addition, one also needs to install PyTorch-Geometric following the official installation instructions.

The code is tested with the following PyTorch-Geometric version.

torch-scatter==2.0.5
torch-sparse==0.6.7
torch-cluster==1.5.7
torch-geometric==1.6.1

Run the code

Example: python main.py --lr 0.001 --hidden_size 16 --dataset wiki-squirrel --model_type regcgcn.

Cite

@article{ma2020copulagnn,
  title={CopulaGNN: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks},
  author={Ma, Jiaqi and Chang, Bo and Zhang, Xuefei and Mei, Qiaozhu},
  booktitle={International Conference on Learning Representations},
  year={2021}
}

About

CopulaGNN: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks (ICLR 2021)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages