Install the dependencies from environment.yml file using the command: conda env create -f environment.yml
Download the following folders before running the models:
- data/ folder from https://drive.google.com/drive/folders/1H8TFDUzIFAezM1rVoB_1PQsCxNyqAQ2f?usp=sharing
- saved_models/ folder from https://drive.google.com/drive/folders/1EQwDw0JQzSXDEN5gaadN3LMxVyjDSYL8?usp=sharing
In src/ folder run train.py for training and eval.py in for evaluation
Config File: config.yml
- epochs: Number of epochs to train
- batch_size: Batch size to use
- model: Which Model to train/evaluate. Can be 'ct', 'cat' or 'mpctm'
Trained models are stored in saved_models/ folder and Evaluation results are stored in results/ folder