- Ubuntu22.04
- Python 3.10
- PyTorch 2.2.0+cu121
- GPU 3090 24G
- RAM 64G
install the requirements
sh GNN/install_requirements.sh
build the graph ,and this may cost lots of time,if need you can find me to get my built graph
sh GNN/build_graph.sh
train and save the model
sh GNN/train_model.sh
predict
python GNN/w2v_model_test.py
python GNN/sci_model_test.py
To perfectly reproduce the effect, you can load the trained model for inference using the following code
python GNN/trained_w2v_model_test.py
python GNN/trained_sci_model_test.py
sh lgb+cat+xgb/install_requires.sh
train_w2v
python lgb+cat+xgb/w2vProcess.py
train-and-predict
python lgb+cat+xgb/main.py
python melt.py
IND Dataset:Data (biendata.xyz)
SciBERT model:allenai/scibert_scivocab_uncased at main (hf-mirror.com)
Trained Model or More externel data:链接:https://pan.baidu.com/s/1toWSZvdgm-ZTEBogzgNJsw 提取码:deep
You can see more details in GNN/README.md and lgb+cat+xgb/README.md
- DeepMayNotLearn, junior student of CQUPT
- yueming, junior student of CQUPT
- yxw666, junior student of CQUPT