python3 process_1.py
python3 process_2.py
python3 process_3.py
python3 process_4.py
where process_1.py,process_2.py are data process. (need 20 minate)
+ process_3.py is the graph construction (need 5 hour)
+ process_4.py is training model and evaluation.
The results will show the performance of RetaGNN-NLP by classification metric Accuracy, Recall, Precision and F1-Score.