This workspace is designed to allow the user to test multiple graph neural network architectures for both regression tasks and classification tasks. The main function of this repo is to allow for that experimentation via a Docker container for rapid development. The secondary goal of this project is to learn how to webscrape for dowloadable files and parse those files for the desired information. This will result in a nearly fully automated pipeline of data acquisition, database creation, and machine learning architecture.
Download the repo
As contributor:
Once gaining contributor permissions,
In terminal with folder open type make to create environment and kernel to have all necessary packages
Do work in notebooks
Please if Python modules are created place them in the correct folders or create a new folder if the correct one does not exist