Chapter 8: TPOT Model Deployment

In this chapter, you'll learn how to deploy any automated machine learning model, both to localhost and the cloud. You'll learn why the deployment step is necessary if you aim to make machine learning-powered software. It's assumed you know how to train basic regression and classification models with TPOT. Knowledge of the topics of the last couple of chapters (Dask and neural networks) isn't required, as we won't deal with those here.

Throughout the chapter, you'll learn how easy it is to wrap your models in an API and expose their predictive capabilities to other users that aren't necessarily data scientists. You'll also learn which cloud providers are the best to get you started entirely for free.

This chapter will cover the following topics:

• Why do we need model deployment?
• Introducing Flask and Flask-RESTful
• Best practices for deploying automated models
• Deploying machine learning...