Relational Temporal Graph Convolutional Network for Ranking-based Stock Prediction
Data and pretained model have been uploaded to Google drive
Place the data on the path: RTGCN/data/
Place the pretained model on the path: RTGCN/work_dir/NASDAQ(OR NYSE)
Tesing pretained model on NASDAQ. Edit the "weights" and "relation_path" variables to reload the other models weights.
python main.py recognition -c config/rt_gcn/NASDAQ/test.yaml
python main.py recognition -c config/rt_gcn/NYSE/test.yaml