The Yolov5s pretained model is deployed using flask. This repo contains example apps for exposing the yolo5 object detection model from pytorch hub via a flask api/app.
Simple app that enables live webcam detection using pretrained YOLOv5s weights and see real time inference result of the model in the browser.
Run locally and dev:
conda create -n <VENV>
conda activate <VENV>
(<VENV>) $ pip install -r requirements.txt
(<VENV>) $ flask run
The example dockerfile shows how to expose the rest API:
# Build
docker build -t yolov5 .
# Run
docker run -p 5000:5000 yolov5-flask:latest