This repository contains a complete setup for training and evaluating a Unified BART model using custom dataset handling and trainer modifications. (The checkpoints and data is not included in the repository yet. It will be released soon.)

• data/: Directory to store datasets.
• models/: Saved model checkpoints and evaluation metrics.
• src/: Source code files including model, dataset, and evaluation scripts.

1. Training the Model To train the model, use the following command:

   python src/train.py

2. Evaluating the Model To evaluate the model, run:

   python src/evaluate.py