Micro service architecture demo using rest API
Currently, we use object detection algorithm for demo, but you can change it as you like.
You can find this project's slide -> Plone Conference 2018: Micro Service Architecture with Machine Learning Application
- REST API server, which is connected to Object Detection server inside it.
- Object Detection server with grpc connection
- A object detection test script
- Python 3.6.6
- ChainerCV 0.8.0
- grpc
- Launch both servers named 'run_api_server.py' and 'run_detection_server.py'
- You can access to the api server bellow command.
$ curl -X POST -F image=@image/dog.jpg 'http://localhost:5000/predict'
# You can get these result.
{"predictions":[{"bbox":{"xmax":"402","xmin":"193","ymax":"388","ymin":"108"},"class":"dog","probability":"0.9997122883796692"}],"success":true}