Object Detection Application with a frontend built using ReactJS, backend built using Flask and Model is based on Tiny-Yolo v3.
The Application has 3 parts:
- client
- server
- yolo model
cd client
npm install
npm start
cd server
pip3 install virtualenv
python -m virtualenv venv
venv/Scripts/activate
deactivate
pip install -r requirements.txt
flask run
$env:FLASK_ENV = "development"
Pull the docker image from dockerhub
create a container:
docker run -p 8501:8501 \
--mount type=bind,source={source path to model},target=/models/yolo
-e MODEL_NAME=yolo -t tensorflow/serving &
the model will be running on http://localhost:8501/v1/models/yolo