Pytorch implementation of HiCAP: Hierarchical Clustering-based Attention Pooling for Graph Representation Learning
A novel method called HiCAP (Hierarchical Clustering-Based Attention Pooling) is proposed to address the limitations of existing graph pooling approaches.
* Python >= 3.6
* Pytorch >= 2.0.1+cu118
* Pytorch_geometric >= 2.4.0
* Numpy
* Networkx
python3 main.py
In case of using CUDA:
CUDA_LAUNCH_BLOCKING=1 python3 main.py
NOTE: there are various arguments which you can modify, please see main.py
for further details.
@INPROCEEDINGS{10326268,
author={Haddadian, Parsa and Abedian, Rooholah and Moeini, Ali},
booktitle={2023 13th International Conference on Computer and Knowledge Engineering (ICCKE)},
title={HiCAP: Hierarchical Clustering-based Attention Pooling for Graph Representation Learning},
year={2023},
volume={},
number={},
pages={463-468},
doi={10.1109/ICCKE60553.2023.10326268}}
Please contact Parsa with any questions.