implementation of paper Unsupervised Attributed Network Embedding via Cross Fusion (WSDM 2021)
- python >= 3.6
- pytorch >= 1.4
- network >= 1.11
For default parameters, use following command
python train.py
If you want to use other parameter setting and dataset, you can use following command for optional arguments description.
python train.py -h
| file | description |
|---|---|
| train.py | The main training code |
| model.py | The implementation of CFANE |
| aggregators.py | Aggregators for information propagation |
| utils.py | Data loading |
| node2vec.py | Generating random walk contexts. Refer to: https://github.com/aditya-grover/node2vec |
We provide Cora dataset and partitions of its ego-network as example of data format.
Our ego-network partition refers to https://github.com/google-research/google-research/tree/master/graph_embedding/persona