SPUG implements data pipeline, data cleaning, graph construction, and modeling with Graph Neural Network.
Start with creating a python virtual environment by following commands
conda create -- name spug python=3.8.5
Install the dependencies by
conda activate spug
./env.sh
The project is driven by configuration file: a yaml file specified every parameters. You can refer configs/example.yaml
for your own configuration file. Our report is mainly investigating dow and jones listing, so our report refers to configs/dow_jones.yaml
.
You may refer the steps in notebooks
to use the project