Full end-to-end deployment of a machine learning algorithm into a live production environment

Click for: original source

Older article will guide you through how to use scikit-learn, pickle, Flask, Microsoft Azure and ipywidgets to fully deploy a Python machine learning algorithm into a live, production environment. By Graham Harrison.

You will get these steps describing how machine learning algorithm could be fully deployed into a live production environment so that it could be “consumed” in a platform-agnostic way:

  • Develop a machine learning algorithm
  • Make an individual prediction from the trained model
  • Develop a web service wrapper
  • Deploy the web service to microsoft Azure
  • Add the Azure app service extension to VS Code
  • Building a client application to consume the Azure deployed web service

There are quite a few steps involved, but using readily available libraries and free tools including scikit-learn, pickle, flask, Microsoft Azure and ipywidgets we have constructed a fully working, publicly available cloud deployment of a machine learning algorithm and a fully functioning client to call and consume the web service and display the results. Nice one!

[Read More]

Tags big-data data-science devops