This is a PyTorch implementation of the Homophily-Enhanced Self-supervision (HES-GSL), and the code includes the following modules:
-
Dataset Loader (Cora, Citeseer, Pubmed, ogbn-arxiv, Texas, Cornell, Wisconsin, Actor, MNIST, and Fashion-MNIST)
-
Various supervision signals
-
Training paradigm pre-training and fine-tuning on ten datasets
-
Visualization and evaluation metrics
- networkx==2.5
- numpy==1.19.2
- scipy==1.3.1
- torch==1.6.0
- dgl==0.6.1
-
main.py
- get_loss_classification() -- Calculate downstream supervision
- get_loss_reconstruction() -- Calculate reconstruction-based self-supervision
- main() -- Pre-train and fine-tune model for node classification task on ten real-world datasets
-
model.py
- GCN_CLA() -- GCN classifier
- GSL() -- Learn a task-specific underlying graph structure
- GCN_DAE() -- Denoising Autoencoder
- get_loss_homophily() -- Calculate the homophily-enhanced self-supervision
-
dataset.py
- load_data() -- Load Cora, Citeseer, Pubmed, Texas, Cornell, Wisconsin, and Actor datasets
- load_ogb_data() -- Load ogbn-arxiv dataset
- load_mnist_data() -- Load MNIST dataset
- load_fashionmnist_data() -- Load Fashion-MNIST dataset
-
utils.py
- get_random_mask() -- Add noise to input node features
- top_k() -- Construct a kNN graph
- knn_fast() -- Construct a kNN graph in a faster manner
- get_homophily() -- Calculate the homophily ratio of the generated graph
The datasets used in this paper are available in:
https://drive.google.com/file/d/1bCkI1fTmgHLHesPkyzUtAIriLPswZZ4f/view?usp=share_link
-
Install the required dependency packages
-
To get the results on a specific dataset, please run with proper hyperparameters:
python main.py --dataset data_name
where the data_name is one of the ten datasets (Cora, Citeseer, Pubmed, ogbn-arxiv, Texas, Cornell, Wisconsin, Actor, MNIST, and Fashion-MNIST). Use the Cora dataset as an example:
python main.py --dataset cora
If you find this project useful for your research, please use the following BibTeX entry.
@article{wu2023beyond,
title={Beyond homophily and homogeneity assumption: Relation-based frequency adaptive graph neural networks},
author={Wu, Lirong and Lin, Haitao and Hu, Bozhen and Tan, Cheng and Gao, Zhangyang and Liu, Zicheng and Li, Stan Z},
journal={IEEE Transactions on Neural Networks and Learning Systems},
year={2023},
publisher={IEEE}
}
Homophily-Enhanced Self-supervision (HES-GSL) is released under the MIT license.