/ml-microservices-kubernetes

This project aims to containerize and operationalize a pre-trained and working sklearn model (that has been trained to predict housing prices in Boston according to several features, such as average rooms in a home and data about highway access, teacher-to-pupil ratios, and so on), machine learning microservice using Kubernetes

Primary LanguagePython

CircleCI

A CircleCI status badge, included above

Project Overview

In this project, we will operationalize a Machine Learning Microservice API.

I have used a pre-trained, sklearn model that has been trained to predict housing prices in Boston according to several features, such as average rooms in a home and data about highway access, teacher-to-pupil ratios, and so on. You can read more about the data, which was initially taken from Kaggle, on the data source site. The aim of this project however, is to operationalize a Python flask app—in a provided file, app.py—that serves out predictions (inference) about housing prices through API calls. This project could be extended to any pre-trained machine learning model, such as those for image recognition and data labeling.

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. In this project I have:

  • Test the project code using linting
  • Complete a Dockerfile to containerize this application
  • Deploy the containerized application using Docker and made a prediction
  • Improve the log statements in the source code for this application
  • Configured Kubernetes and create a Kubernetes cluster
  • Deployed a container using Kubernetes and made a prediction
  • Uploaded a complete Github repo with CircleCI to indicate that the code has been tested

The final implementation of the project will showcase one's abilities to operationalize production microservices.

Setup the Environment

  • Clone the repository
  • Create a virtualenv with Python 3.7 and activate it.
python3 -m pip install --user virtualenv
# You should have Python 3.7 available in your host. 
# Check the Python path using `which python3`
# Use a command similar to this one:
python3 -m virtualenv --python=<path-to-Python3.7> .devops
source .devops/bin/activate
# Complete Makefile as needed to install the dependencies in the requirements.txt file
  • Run make install to install the necessary dependencies

Running app.py

  1. Standalone: python app.py
  2. Run in Docker: ./run_docker.sh
  3. Upload Image to DockerHub: ./run_docker.sh
  4. Run in Kubernetes: ./run_kubernetes.sh

Kubernetes Steps

  • Setup and Configure Docker locally
  • Setup and Configure Kubernetes locally (Using Docker as driver to run Minikube)
  • Create Flask app in Container *docker run -p 8000:80 max-app:v1.0 *docker ps -- to view running containers.
  • Run via kubectl
    • kubectl run max-kube-app --image=$dockerpath --port=80
    • kubectl get pods --to see the list of running pods *kubectl port-forward max-kube-app 8000:80 -- to forward the ports

Folder and File Structures

  • app.py: our flask application file
  • Makefile: Make utility file which defines set of tasks to be executed
  • Dockerfile: file with instructions to build Docker images
  • requirements.txt: file with listed project dependecies
  • docker_out.txt and kubernetes_out.txt:These are the output files of running both the predictions in Docker and Kubernetes.
  • make_predictions.sh, run_docker.sh, run_kubernetes.sh, upload_docker.sh: These are the script files with a sequence of commands ran to perform a given task.