jxy d77b4795ac | 1 month ago | |
---|---|---|
.. | ||
README.md | 1 year ago | |
graphgan_trainer.py | 1 month ago |
Dataset | # Nodes | # Edges |
---|---|---|
arXiv-GrQc | 5242 | 14496 |
gan_datasets : folds to save training data
The data should be an undirected graph in which node IDs start from 0 to N-1 (N is the number of nodes in the graph). Each line contains two node IDs indicating an edge in the graph.
txt file sample :
0 1
3 2
...
Note: the dimension of pre-trained node embeddings should equal parameter 'n_emb'
gan_cache
gan_results
checkpoint : folds to save the model weights
TL_BACKEND="paddle" python graphgan_trainer.py --batch_size_dis 1024 --batch_size_gen 1024 --n_epochs 30 --lr_gen 1e-5 --lr_dis 1e-5
TL_BACKEND="tensorflow" python graphgan_trainer.py --batch_size_dis 1024 --batch_size_gen 1024 --n_epochs 30 --lr_gen 1e-5 --lr_dis 1e-5
TL_BACKEND="torch" python graphgan_trainer.py --batch_size_dis 1024 --batch_size_gen 1024 --n_epochs 30 --lr_gen 1e-5 --lr_dis 1e-5
Dataset | Paper | Our(pd) | Our(tf) | Our(torch) |
---|---|---|---|---|
arXiv-GrQc | 0.849 | 0.8813±0.00069 | 0.8819±0.00133 | 0.8820 |
No Description
Python C++ Cuda Markdown Text
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》