Project Overview

This project consists in operationalizing a Machine Learning Microservice API based on a Python flask-app.

Project Tasks

The project goal is to operationalize this working, machine learning microservice using kubernetes, which is an open-source system for automating the management of containerized applications. This project will cover:

• Testing the code using linting
• Completing a Dockerfile to containerize this application
• Deploying the containerized application using Docker and making a prediction
• Improving the log statements in the source code for this application
• Configuring Kubernetes and creating a Kubernetes cluster
• Deploying a container using Kubernetes and making a prediction
• Uploading a complete Github repo with CircleCI to indicate that the code has been tested

---

Setup the Environment

• Create a virtualenv and activate it
• Run make install to install the necessary dependencies

Running app.py

1. Standalone: python app.py
2. Run in Docker: ./run_docker.sh
3. Run in Kubernetes: ./run_kubernetes.sh

Uploading the Docker image

To upload the image to Docker Hub: ./upload_docker.sh

Note that running this script will prompt for the username and password of the Docker account.

Kubernetes Steps

• Setup and Configure Docker locally

• Setup and Configure Kubernetes locally

  • minikube start to start a local cluster
  • kubectl config view to check that the cluster has been started (there should be least one cluster with a certificate-authority and server)

• Create Flask app in Container and run via kubectl ./run_kubernetes.sh

• After you’re are done deploying the containerized application and making test predictions via Kubernetes cluster, you should clean up your resources and delete the kubernetes cluster with a call to minikube delete

Files

• output_txt_files/docker_out.txt containes the prediction returned after running the app with Docker

• output_txt_files/kubernetes_out.txt containes the prediction returned after running the app with Kubernetes

• run_docker.sh contains the steps to get Docker running the app locally:

  • Build the image and add a descriptive tag
  • Run flask app

• run_kubernetes.sh contains the steps to get Kubernetes running the app locally:

  • Run the Docker Hub container with kubernetes
  • Forward the container port to a host

• upload_docker.sh contains the steps to upload the image to the Docker repository:

  • Authenticate & tag the image
  • Push image to a docker repository