Skip to content

krishnaadithya/retfound_api

Repository files navigation

Retfound Model API

This repository contains the Retfound model fine-tuned on the IDRID dataset and an API built with Flask to host and interact with the model.

Install Environment

  1. Create a Python environment with conda:

    conda create -n retfound python=3.7.5 -y
    conda activate retfound
  2. Install dependencies:

    git clone https://github.com/krishnaadithya/retfound_api.git
    cd retfound_api
    pip install -r requirements.txt

Steps to Use the API

1. Download Model File

  • Find the model .pth file here.
  • Download the file and place it in the finetune_IDRID folder in the root directory.

2. Running the API

  • The API is built using Flask.

  • Host the model by running:

    python retfound_api.py

3. Accessing the API

  • For online tunneling, the current setup uses ngrok.

  • Currently, the model is hosted locally, and you can use the following endpoint to push your image: https://623c-2001-8f8-166b-2a54-b5d7-fefa-f504-a420.ngrok-free.app/predict.

  • Api file format: {'image': (image_path, image_data)}

  • Use the provided Python code to call the API:

    import requests
    from PIL import Image
    import io
    import time
    
    # Replace 'your_image.jpg' with the path to your image file
    image_path = 'your_image.jpg'
    
    # Open the image file
    with open(image_path, 'rb') as f:
        image_data = f.read()
    
    # Create a dictionary containing the image file
    files = {'image': (image_path, image_data)}
    
    # Make a POST request to the API endpoint
    url = "https://623c-2001-8f8-166b-2a54-b5d7-fefa-f504-a420.ngrok-free.app/predict"  # Replace this link with your endpoint
    response = requests.post(url, files=files)
    
    
    # Check if the request was successful (status code 200)
    if response.status_code == 200:
        result = response.json()
        print(f'Predicted category: {result["predictions"]}')    
    else:
        print(f'Error: {response.text}')

Please replace your_image.jpg with the path to your image file before running the code.