Assignment 4 for ANLP 11-711 -- Group 41

The repo is initially forked from our assignment 3 repo. We resue some training code from assignment 3, and add code for data augmentation as well as transfer learning.

Usage:

• pip install -r requirements.txt
• run accelerate config to configure your accelerate settings
• For transfer learning, run accelerate launch {target_file}.py
• To replicate our augmented datasets, cd to 11711_assignment_4/code/data_aug folder, choose an augmentation method and run the corresponding notebook step by step