Flask REST API

REST APIs are commonly used to expose Machine Learning (ML) models to other services. This folder contains an example REST API created using Flask to expose the YOLOv5s model from PyTorch Hub.

Requirements

Install requirements.txt in a Python>=3.7.0 environment, including PyTorch>=1.7.

$ pip install -r requirements.txt

Run

After Flask installation run:

$ export FLASK_ENV=development && flask run

or

$ python app.py

To use custom model:

$ python app.py --model best.pt

Use GUI

Use GUI to upload and view resut: Go to http://localhost:5000

Use Terminal

Use curl to perform a request (Recommended: Git Bash):

$ curl -X POST -F image=@images/cat.jpg 'http://localhost:5000/v1/object-detection/yolov5s'

The model inference results are returned as a JSON response:

[{"xmin":384.2033081055,"ymin":35.8358840942,"xmax":823.404296875,"ymax":409.2363891602,"confidence":0.8502380848,"class":0,"name":"cat"}]

Example

Reference

• YOLOv5 Flask REST API
• Yolov5 Flask @jzhang533
• Load YOLOv5 from PyTorch Hub
• Pytorch Flask API Heroku @avinassh

Note: The best.pt provided is self-trained yolov5 model on cats and dogs dataset.