Change detection repo, contains the codebase used to obtain results so far

Setting up the environment

1. conda create -n FinderV2 python=3.10
2. conda activate FinderV2
3. pip install -r requirements.txt

Run sequence

1. python data_assoc.py
2. python choose_images.py
3. python data.py to create a train/val/test split for the images
4. python lora.py for training LoRA ViT and showing results