ml_model_server_docker_deployment

This repository is for containing source codes of machine learning model server deployment.

Prerequisite

You need to have docker

Youtube Video of The Project

Docker Deployment Instructions

1. clone this repo to your local machine

2. open terminal

3. cd /path/to/ml_model_server_docker_deployment

4. start docker

5. use the following command to build the docker image

docker image build -t docker-iris-ml-server .

6. use the following command to see docker images

docker image ls

7. use the following command to run ml server

docker run -p 5000:5000 -d docker-iris-ml-server

8. open http://localhost:5000/predict_class in your browser to see get method message

9. open postman and try http://localhost:5000/predict_class for the post method to get model prediction

For example; try the following in your POST request to get machine learning model prediction

{"sepal_length": 5.1, "sepal_width": 3.5, 
"petal_length": 1.4, "petal_width": 0.2}

10. use the following command to see CONTAINER ID

docker ps

11. copy CONTAINER ID

12. use the following command to stop docker container

docker stop "CONTAINER ID"

13. use the following command to free up all the resources that the docker container uses

docker system prune