flask yolov5 code

Sure! Here's an example code snippet showing how to integrate YOLOv5 with Flask:

from flask import Flask, request, jsonify
from PIL import Image
import torch

app = Flask(__name__)

# Load YOLOv5 model
model = torch.hub.load('ultralytics/yolov5', 'yolov5s', pretrained=True).autoshape()

@app.route('/detect', methods=['POST'])
def detect_objects():
    # Get image from request
    image_file = request.files['image']
    image = Image.open(image_file).convert('RGB')

    # Perform object detection
    results = model(image)

    # Extract bounding box coordinates and class labels
    boxes = results.xyxy[0].tolist()
    labels = results.names[0]

    # Prepare response
    response = {
        'boxes': boxes,
        'labels': labels
    }

    return jsonify(response)

if __name__ == '__main__':
    app.run()

This code sets up a Flask server that listens for POST requests on the `/detect` endpoint. It expects an image file in the request payload. The YOLOv5 model is loaded using `torch.hub.load()` and performs object detection on the input image. The detected bounding box coordinates and class labels are then returned as a JSON response.

Note: Make sure you have the necessary dependencies installed, such as Flask, PIL, and torch. You can install them using pip:

pip install flask pillow torch torchvision

Remember to customize and modify the code according to your specific requirements and file paths.