A PyTorch implementation of NA-GSL: Exploring attention mechanism for graph similarity learning (Knowledge-Based Systems) paper link
We propose a unified graph similarity learning framework, Node-wise Attention guided Graph Similarity Learning, NA-GSL, involving i) a hybrid of graph convolution and graph self-attention for node embedding learning, ii) a cross-attention (GCA) module for graph interaction modeling, iii) similarity-wise self-attention (SSA) module for graph similarity matrix fusion and alignment and iv) graph similarity structure learning for predicting the similarity score.
- python==3.8
- pytorch==1.10.2
- torch_geometric==1.10
- tqdm
- scipy
- texttable
cd src
python main.py --dataset=LINUX