Awesome

RTGCN

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

Testing pretained model on NYSE

python main.py recognition -c config/rt_gcn/NYSE/test.yaml