Repo Organization:
Code: contains all code files for the project, including train script, test script, and interpretability script. To run in a Linux terminal, add the argument --path <CLONED_GIT_DIRECTORY>. Defaults in the files should be left as-is to recreate our results, however toggling off the code to save figures, models, and data files will not affect the models but may cause subsequent processes to break.
- To re-train the model from scratch, run model_train.py
- To run testing only, using current model, run model_test.py
Data: intermediate repository for data. Initial data grab is from Google Drive.
Final-Group-Presentation: contains final .pdf copy of the group's presentation
Final-Group-Project-Report: contains final .pdf of the group's report
Group-Proposal: Contains initial rough proposal for project (note: original proposal differs from the final project. Ideas don't always work.)