To run the code, we need to create a virtual environment using Anaconda, and install the required dependencies.The command is as follows：
conda create -n ProLoc-IHS pyhton=3.7.13
conda activate ProLoc-IHS
pip install -r requirements.txt
In the dataset folder, there are four csv files, where train/test.csv is the training set and test set information, including labels, sequences and other information, and train/test_img_URL.csv contains the URL address of the image information. You need to use python download.py to download the IHC image.
We use a pre-trained language model to extract features. You need to download the ProtT5([website]) and Vit(download it from Huggingface) models to your local computer.And Then run
python feature extraction.py
After preparing the features and training files,run:
python train.py
and
python test.py