Parser

Multi-level Disentanglement Graph Neural Network

This is a PyTorch implementation of the MD-GNN, and the code includes the following modules:

  • Datasets (Cora, Citeseer, Pubmed, Synthetic, and ZINC)

  • Training paradigm for node classification, graph classification, and graph regression tasks

  • Visualization

  • Evaluation metrics

Main Requirements

  • dgl==0.4.3.post2
  • networkx==2.4
  • numpy==1.18.1
  • ogb==1.1.1
  • scikit-learn==0.22.2.post1
  • scipy==1.4.1
  • torch==1.5.0

Description

  • train.py

    • main() — Train a new model for node classification task on the Cora, Citeseer, and Pubmed datasets
    • evaluate() — Test the learned model for node classification task on the Cora, Citeseer, and Pubmed datasets
    • main_synthetic() — Train a new model for graph classification task on the Synthetic dataset
    • evaluate_synthetic() — Test the learned model for graph classification task on the Synthetic dataset
    • main_zinc() — Train a new model for graph

       

       

       

      To finish reading, please visit source site