Add demo UI, token attention rollout & top5 table; clean ignores

fix model paths, retrain, perform inference

modify files to accomodate inference script

update test csv file path

script for generating test data

train image only model

modify path in train script

modify train.py for wandb

Resolve paths to images

Add new tests

Add changes for CI complicance

add new tests and fixed paths

added safeguards

fixed csv path in generate_emr_csv.py

Auto-format code with Black

formatting changes

Add tests, formatting, and CI enhancements with pre-commit support

fixed formatting

made formatting changes for CI pipeline

Made changes to optimize hyperparams for ablation study

made directory changes, after findind the optimal data generation script

experimenting with data to make it more realistic

trained different models and plotted accuracy curves

changes to incorporate text only, image only model modes

made changes to dataset script to do text-only or image-only training

hyper parameter tuning complete best params in train.py

made changes to avoid overfitting

modified data generator and optuna trainer to prevent overfitting

hyperparameter tuning with optuna, wandb

first version of model training script

created a dataset for csvfiles, images

added init.py in src to make it a pkg, made changes in multimodal_model.py

Created a multimodal model to combine text and images

made path changes to generate_emr, tested using data_preprocessing

Added Script to generate emr data in csv