Operationalizing Machine Learning Pipelines: Building Reusable and Reproducible Machine Learning Pipelines Using Mlops

This book will provide you with an in-depth understanding of MLOps and how you can use it inside an enterprise. Each tool discussed in this book has been thoroughly examined, providing examples of how to install and use them, as well as sample data.

This book will teach you about every stage of the machine learning lifecycle and how to implement them within an organisation using a machine learning framework. With GitOps, you'll learn how to automate operations and create reusable components such as feature stores for use in various contexts. You will learn to create a server-less training and deployment platform that scales automatically based on demand. You will learn about Polyaxon for machine learning model training, and KFServing, for model deployment. Additionally, you will understand how you should monitor machine learning models in production and what factors can degrade the model's performance.

You can apply the knowledge gained from this book to adopt MLOps in your organisation and tailor the requirements to your specific project. As you keep an eye on the model's performance, you'll be able to train and deploy it more quickly and with greater confidence.

TABLE OF CONTENTS

1. DS/ML Projects - Initial Setup

2. ML Projects Lifecycle

3. ML Architecture - Framework and Components

4. Data Exploration and Quantifying Business Problem

5. Training & Testing ML model

6. ML model performance measurement

7. CRUD operations with different JavaScript frameworks

8. Feature Store

9. Building ML Pipeline