Source code (PyTorch) and dataset of the paper "Adapting Precomputed Features for Efficient Graph Condensation", which is accepted by Forty-second International Conference on Machine Learning (ICML 2025).
- Implementation: https://github.com/Xtra-Computing/GCPA
- Paper Access: https://openreview.net/forum?id=ThK6o74QLc
I have tested this environment with the following requirements, but please note that you are not restricted to these specific versions and can choose other versions as appropriate:
- python=3.10.13
- torch=2.4.1
- dgl=2.4.0
- ogb=1.3.6
- numpy=1.26.4
- scikit-learn=1.6.1
- wandb=0.19.7
./run.sh
If you use GCPA in a scientific publication, we would appreciate citations to the following paper:
@inproceedings{
li2025adapting,
title={Adapting Precomputed Features for Efficient Graph Condensation},
author={Yuan Li and Jun Hu and Zemin Liu and Bryan Hooi and Jia Chen and Bingsheng He},
booktitle={Forty-second International Conference on Machine Learning},
year={2025},
url={https://openreview.net/forum?id=ThK6o74QLc}
}
License: GPLv3
Copyright (c) 2024-2025 Xtra Computing Group, NUS, Singapore.