- Install Python in at least 3.10 version
- Install required packages:
-
pip install -r requirements.txt
- To use the CPU or GPU version, install the appropriate version of the PyTorch library according to the instructions: https://pytorch.org/get-started/locally/#windows-pip
- Download the trained models from the following address: https://drive.google.com/drive/folders/1ttZ6TEqxMglzeY46YM0IWJwrV0o8g97V?usp=sharing
- Extract the models to the models directory in the main solution path.
- In line 9 of the main.py file, set the path to the directory containing the fundus images that will be processed by the method.
- On line 97, enter the file extension for the fundus photos.
- Run the main.py file.