Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
Guangyu Zhou eaa51ac4dc | 3 months ago | |
---|---|---|
.. | ||
iehgcn_trainer.py | 3 months ago | |
readme.md | 10 months ago |
Dataset | # Nodes | # Node Types | # Edges | # Edge Types | Target | # Classes |
---|---|---|---|---|---|---|
DBLP | 26,128 | 4 | 239,566 | 6 | author | 4 |
IMDB | 21,420 | 4 | 86,642 | 6 | movie | 4 |
DBLP dataset refer to HGBDataset.
IMDBdataset refer to IMDB.
For the DBLP dataset: train test val = 974, 1420, 243 about 37% for training.
For the IMDB dataset: train test val = 400, 3478, 400, about 9% for training.
Dataset | Paper(80% training) | Paper(60% training) | Paper(40% training) | Paper(20% training) | Our(tf) | Our(th) | Our(pd) |
---|---|---|---|---|---|---|---|
DBLP | 96.29 | 95.25 | 93.83 | 93.85 | 92.30±0.49% | 90.90±0.74% | 91.18±0.66% |
IMDB | 58.35 | 60.84 | 59.81 | 56.60 | 58.10±0.42% | 55.22±1.21% | 56.08±2.13% |
TL_BACKEND="tensorflow" python3 iehgcn_trainer.py --dataset DBLP --n_epoch 30 --lr 0.01 --num_layers 3 --hidden_channels [64, 32] --l2_coef 0.0005 --drop_rate 0.2
TL_BACKEND="torch" python3 iehgcn_trainer.py --dataset DBLP --n_epoch 30 --lr 0.005 --num_layers 4 --hidden_channels [64, 32, 16] --l2_coef 0.0005 --drop_rate 0.0
TL_BACKEND="paddle" python3 iehgcn_trainer.py --dataset DBLP --n_epoch 30 --lr 0.01 --num_layers 4 --hidden_channels [64, 32, 16] --l2_coef 0.0005 --drop_rate 0.1
TL_BACKEND="torch" python3 iehgcn_trainer.py --dataset IMDB --n_epoch 25 --lr 0.01 --num_layers 3 --hidden_channels [64, 32] --l2_coef 0.0005 --drop_rate 0.2
TL_BACKEND="tensorflow" python3 iehgcn_trainer.py --dataset IMDB --n_epoch 25 --lr 0.005 --num_layers 3 --hidden_channels [64, 32] --l2_coef 0.0005 --drop_rate 0.2
TL_BACKEND="paddle" python3 iehgcn_trainer.py --dataset IMDB --n_epoch 25 --lr 0.005 --num_layers 3 --hidden_channels [64, 32] --l2_coef 0.0005 --drop_rate 0.2
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》