Codes for the paper:《Robust Network Alignment with the Combination of Structure and Attribute Embeddings》.
pytorch 1.10.2,networkx 2.3,python 3,cuda 10.2,dgl 1.1.0+cu102
- algorithms
- GATv2:contains the embedding model:
multi_GATv2.py
andembedding.py
, the final run filerun_method.py
- refine: code for topological refinement
- GATv2:contains the embedding model:
- evaluation: contains the evaluation methods for the models
- graph data: contain
Douban
andPPI
datasets. - input: to generate the graph data
- utils: contains some files for add noise to graph and generate target graphs for synthetic datas.
Go to MyAlign\algorithms\MyMethod\GATv2\run_method.py
, and choose the data you want to test to run.