Theheavens 37cf43fe68 | 2 years ago | |
---|---|---|
.. | ||
README.md | 2 years ago |
Paper: Heterogeneous Graph Structure Learning for Graph Neural Networks
Code from author:
https://github.com/Andy-Border/HGSL
Clone the Openhgnn-DGL
python main.py -m HGSL -d acm4GTN -t node_classification -g 0 --use_best_config
If you do not have gpu, set -gpu -1.
Node classification
Node classification | acm4GTN macro-f1 | acm4GTN micro-f1 |
---|---|---|
paper | 93.48 | 93.37 |
OpenHGNN | 93.28 | 93.18 |
The model is trained in semi-supervisied node classification.
Supported dataset: acm4GTN
We process the acm4GTN dataset with adding the metapath2vec embeddings obtained from the dataset of the author's code.
Requirements for datasets
hidden_dim = 16
num_heads = 2
gnn_emd_dim = 64
The best config for each dataset can be found in best_config.
This model under the best config has some slight differences compared with the code given by the paper author,which seems having little impact on performance:
Xinlong Zhai[GAMMA LAB]
Submit an issue or email to zhaijojo@bupt.edu.cn.
OpenHGNN是由北邮GAMMA Lab开发的基于PyTorch和DGL的开源异质图神经网络工具包。
Python Markdown Shell
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》