Operationalize a Production Microservice

This project is about operationalizing a machine learning microservice inference API app using Docker, Kubernetes, and CircleCI. The microservice app is a Flask based app that serves out predictions or inferences about housing prices through API calls. Furthermore, the app uses an sklearn machine learning training model.

Environment Setup

• Create a Python virtual environment:

      $ python3 -m venv ~/.microservice-inference-api
  

• Activate the virtual environment:

      $ source ~/.microservice-inference-api/bin/activate
  

• Run make install to install the necessary dependencies:

      $ make install
  

Running app.py

1. Standalone:

       $ python app.py
   

2. Run in Docker:

       $ ./run_docker.sh
   

3. Run in Kubernetes:

       $ ./run_kubernetes.sh
   

Kubernetes Steps

• Setup and Configure Docker locally
• Setup and Configure Kubernetes locally
• Create Flask app in Container
• Run via kubectl

List of Some Files in the Repo

• app.py : The Python flask app to operationalize

• Dockerfile : The Docker config file needed to containerize the app

• Makefile : A bash script file used to install application dependencies

• make_prediction.sh : A script file used to make predictions

• model_data : A folder containing housing data information in the Boston area

• output_txt_files : A folder containing the result of the prediction

• requirements.txt : A text file containing a list of Python app dependencies

• run_docker.sh : A script file used to build the Docker image

• run_kubernetes.sh : A script file used to build the Kubernetes container

• upload_docker.sh : A script file used to push the Docker image to the remote Docker repository

• .circleci : - A folder that contains CircleCI's config.yml file that checks the project code for errors

License

The contents of this repository are covered under the MIT Licence.