Book Image

Practical Automated Machine Learning Using H2O.ai

By : Salil Ajgaonkar
Book Image

Practical Automated Machine Learning Using H2O.ai

By: Salil Ajgaonkar

Overview of this book

With the huge amount of data being generated over the internet and the benefits that Machine Learning (ML) predictions bring to businesses, ML implementation has become a low-hanging fruit that everyone is striving for. The complex mathematics behind it, however, can be discouraging for a lot of users. This is where H2O comes in – it automates various repetitive steps, and this encapsulation helps developers focus on results rather than handling complexities. You’ll begin by understanding how H2O’s AutoML simplifies the implementation of ML by providing a simple, easy-to-use interface to train and use ML models. Next, you’ll see how AutoML automates the entire process of training multiple models, optimizing their hyperparameters, as well as explaining their performance. As you advance, you’ll find out how to leverage a Plain Old Java Object (POJO) and Model Object, Optimized (MOJO) to deploy your models to production. Throughout this book, you’ll take a hands-on approach to implementation using H2O that’ll enable you to set up your ML systems in no time. By the end of this H2O book, you’ll be able to train and use your ML models using H2O AutoML, right from experimentation all the way to production without a single need to understand complex statistics or data science.

Related Content you might be interested in

Current Title:

Small book image
Practical Automated Machine Learning Using H2O.ai
Salil Ajgaonkar
Sep, 2022 | 396 pages
Small book image
Machine Learning at Scale with H2O
David Whiting | Gregory Keys
Jul, 2022 | 396 pages
Small book image
Ensemble Machine Learning Cookbook
Vijayalakshmi Natarajan | Dipayan Sarkar
Jan, 2019 | 336 pages
Small book image
Mastering Machine Learning with Spark 2.x
Alex Tellez | Max Pumperla | Michal Malohlava
Aug, 2017 | 340 pages
Table of Contents (19 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Download the example code files
Download the color images
Conventions used
Get in touch
Share Your Thoughts
1
Part 1 H2O AutoML Basics
Part 1 H2O AutoML Basics
Free Chapter
2
Chapter 1: Understanding H2O AutoML Basics
Chapter 1: Understanding H2O AutoML Basics
Technical requirements
Understanding AutoML and H2O AutoML
Minimum system requirements to use H2O AutoML
Installing Java
Basic implementation of H2O using Python
Basic implementation of H2O using R
Training your first ML model using H2O AutoML
Summary
3
Chapter 2: Working with H2O Flow (H2O’s Web UI)
Chapter 2: Working with H2O Flow (H2O’s Web UI)
Technical requirements
Understanding the basics of H2O Flow
Working with data functions in H2O Flow
Working with model training functions in H2O Flow
Working with prediction functions in H2O Flow
Summary
4
Part 2 H2O AutoML Deep Dive
Part 2 H2O AutoML Deep Dive
5
Chapter 3: Understanding Data Processing
Chapter 3: Understanding Data Processing
Technical requirements
Reframing your dataframe
Handling missing values in the dataframe
Manipulating feature columns of the dataframe
Tokenization of textual data
Encoding data using target encoding
Summary
6
Chapter 4: Understanding H2O AutoML Architecture and Training
Chapter 4: Understanding H2O AutoML Architecture and Training
Observing the high-level architecture of H2O
Learning about the flow of interaction between the client and the H2O service
Understanding how H2O AutoML performs hyperparameter optimization and training
Summary
7
Chapter 5: Understanding AutoML Algorithms
Chapter 5: Understanding AutoML Algorithms
Understanding the different types of ML algorithms
Understanding the Generalized Linear Model algorithm
Understanding the Distributed Random Forest algorithm
Understanding the Gradient Boosting Machine algorithm
Understanding what is Deep Learning
Summary
8
Chapter 6: Understanding H2O AutoML Leaderboard and Other Performance Metrics
Chapter 6: Understanding H2O AutoML Leaderboard and Other Performance Metrics
Exploring the H2O AutoML leaderboard performance metrics
Exploring other model performance metrics
Summary
9
Chapter 7: Working with Model Explainability
Chapter 7: Working with Model Explainability
Technical requirements
Working with the model explainability interface
Exploring the various explainability features
Summary
10
Part 3 H2O AutoML Advanced Implementation and Productization
Part 3 H2O AutoML Advanced Implementation and Productization
11
Chapter 8: Exploring Optional Parameters for H2O AutoML
Chapter 8: Exploring Optional Parameters for H2O AutoML
Technical requirements
Experimenting with parameters that support imbalanced classes
Experimenting with parameters that support early stopping
Experimenting with parameters that support cross-validation
Summary
12
Chapter 9: Exploring Miscellaneous Features in H2O AutoML
Chapter 9: Exploring Miscellaneous Features in H2O AutoML
Technical requirements
Understanding H2O AutoML integration in scikit-learn
Understanding H2O AutoML event logging
Summary
13
Chapter 10: Working with Plain Old Java Objects (POJOs)
Chapter 10: Working with Plain Old Java Objects (POJOs)
Technical requirements
Introduction to POJOs
Extracting H2O models as POJOs
Using a H2O model as a POJO
Summary
14
Chapter 11: Working with Model Object, Optimized (MOJO)
Chapter 11: Working with Model Object, Optimized (MOJO)
Technical requirements
Understanding what a MOJO is
Extracting H2O models as MOJOs
Viewing model MOJOs
Using H2O AutoML model MOJOs to make predictions
Summary
15
Chapter 12: Working with H2O AutoML and Apache Spark
Chapter 12: Working with H2O AutoML and Apache Spark
Technical requirements
Exploring Apache Spark
Exploring H2O Sparkling Water
Summary
16
Chapter 13: Using H2O AutoML with Other Technologies
Chapter 13: Using H2O AutoML with Other Technologies
Technical requirements
Using H2O AutoML and Spring Boot
Using H2O AutoML and Apache Storm
Summary
17
Index
Index
Why subscribe?
18
Other Books You May Enjoy
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts

Minimum system requirements to use H2O AutoML

H2O is very easy to install, but certain minimum standard requirements need to be met for it to run smoothly and efficiently. The following are some of the minimum requirements needed by H2O in terms of hardware capabilities, along with other software support:

  • The minimum hardware required by H2O is as follows:
    • Memory: H2O runs on an in-memory architecture, so it is limited by the physical memory of the system that uses it. Thus, to be able to process huge chunks of data, the more memory the system, has the better.
    • Central Processing Unit (CPU): By default, H2O will use the maximum available CPUs of the system. However, at a minimum, it will need 4 CPUs.
    • Graphical Processing Unit (GPU): GPU support is only available for XGBoost models in AutoML if the GPUs are NVIDIA GPUs (GPU Cloud, DGX Station, DGX-1, or DGX-2) or if it is a CUDA 8 GPU.
  • The operating systems that support H2O are as follows:
    • Ubuntu 12.04
    • OS X 10.9 or later
    • Windows 7 or later
    • CentOS 6 or later
  • The programming languages that support H2O are as follows:
    • Java: Java is mandatory for H2O. H2O requires a 64-bit JDK to build H2O and a 64-bit JRE to run its binary:
      • Java versions supported: Java SE 15, 14, 13, 12, 11, 10, 9, and 8
    • Other Languages: The following languages are only required if H2O is being run in those environments:
      • Python 2.7.x, 3.5.x, or 3.6.x
      • Scala 2.10 or later
      • R version 3 or later
  • Additional requirements: The following requirements are only needed if H2O is being run in these environments:
    • Hadoop: Cloudera CDH 5.4 or later, Hortonworks HDP 2.2 or later, MapR 4.0 or later, or IBM Open Platform 4.2
    • Conda: 2.7, 3.5, or 3.6
    • Spark: Version 2.1, 2.2, or 2.3

Once we have a system that meets the minimum requirements, we need to focus on H2O’s functional dependencies on other software. H2O has only one dependency and that is Java. Let’s see why Java is important for H2O and how we can download and install the correct supported Java version.

Previous Section
End of Section 4
Next Section