You're reading from Machine Learning at Scale with H2O A practical guide to building and deploying machine learning models on enterprise systems

Product type Paperback

Published in Jul 2022

Publisher Packt

ISBN-13 9781800566019

Length 396 pages

Edition 1st Edition

Tools

H2O

Concepts

Machine Learning

Authors (2):

Gregory Keys

David Whiting

View More author details

Table of Contents (22) Chapters

Preface

1. Section 1 – Introduction to the H2O Machine Learning Platform for Data at Scale

2. Chapter 1: Opportunities and Challenges FREE CHAPTER

3. Chapter 2: Platform Components and Key Concepts

4. Chapter 3: Fundamental Workflow – Data to Deployable Model

5. Section 2 – Building State-of-the-Art Models on Large Data Volumes Using H2O

6. Chapter 4: H2O Model Building at Scale – Capability Articulation

7. Chapter 5: Advanced Model Building – Part I

8. Chapter 6: Advanced Model Building – Part II

9. Chapter 7: Understanding ML Models

10. Chapter 8: Putting It All Together

11. Section 3 – Deploying Your Models to Production Environments

12. Chapter 9: Production Scoring and the H2O MOJO

13. Chapter 10: H2O Model Deployment Patterns

14. Section 4 – Enterprise Stakeholder Perspectives

15. Chapter 11: The Administrator and Operations Views

16. Chapter 12: The Enterprise Architect and Security Views

17. Section 5 – Broadening the View – Data to AI Applications with the H2O AI Cloud Platform

18. Chapter 13: Introducing H2O AI Cloud

19. Chapter 14: H2O at Scale in a Larger Platform Context

20. Other Books You May Enjoy

Appendix : Alternative Methods to Launch H2O Clusters

Feature engineering options

In this section, we will demonstrate how feature engineering can lead to better predictive models. Second only to data cleaning, typically, feature engineering is the most time-consuming of all tasks involved in the modeling process. It can also be the "secret sauce" that makes for a great predictive model.

So, what does feature engineering mean? Put simply, it is how to extract information from raw data into a form that is both usable by the modeling algorithm and interpretable for the problem at hand. For example, a date or date-time object might be represented in data as a string or a number (for example, Unix time is the number of seconds since 00:00:00 UTC on January 1, 1970). Presented with such features, an algorithm is liable to treat dates as levels of a categorical variable or a continuous numeric value. Neither of these forms is very helpful. However, embedded in this raw data is information about not only the day, the month...

The rest of the chapter is locked

You're reading from Machine Learning at Scale with H2O A practical guide to building and deploying machine learning models on enterprise systems

Table of Contents (22) Chapters

Feature engineering options

Unlock this book and the full library FREE for 7 days

Authors (2)

Personalised recommendations for you