Kumar - Machine learning quick reference: quick and essential machine learning hacks for training smart data models

Copyright 2019 Packt Publishing

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews.

Every effort has been made in the preparation of this book to ensure the accuracy of the information presented. However, the information contained in this book is sold without warranty, either express or implied. Neither the author, nor Packt Publishing or its dealers and distributors, will be held liable for any damages caused or alleged to have been caused directly or indirectly by this book.

Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot guarantee the accuracy of this information.

Commissioning Editor: Amey Varangaonkar
Acquisition Editor: Porous Godhaa
Content Development Editor: Ronnel Mathew
Technical Editor: Sagar Sawant
Copy Editor: Safis Editing
Project Coordinator: Namrata Swetta
Proofreader: Safis Editing
Indexer: Priyanka Dhadke
Graphics: Jisha Chirayil
Production Coordinator: Shraddha Falebhai

First published: January 2019

Production reference: 1310119

Published by Packt Publishing Ltd.
Livery Place
35 Livery Street
Birmingham
B3 2PB, UK.

ISBN 978-1-78883-057-7

www.packtpub.com

Mapt is an online digital library that gives you full access to over 5,000 books and videos, as well as industry leading tools to help you plan your personal development and advance your career. For more information, please visit our website.

• Spend less time learning and more time coding with practical eBooks and Videos from over 4,000 industry professionals

• Improve your learning with Skill Plans built especially for you

• Get a free eBook or video every month

• Mapt is fully searchable

• Copy and paste, print, and bookmark content

Did you know that Packt offers eBook versions of every book published, with PDF and ePub files available? You can upgrade to the eBook version at www.packt.com and as a print book customer, you are entitled to a discount on the eBook copy. Get in touch with us at customercare@packtpub.com for more details.

At www.packt.com , you can also read a collection of free technical articles, sign up for a range of free newsletters, and receive exclusive discounts and offers on Packt books and eBooks.

Rahul Kumar has got more than 10 years of experience in the space of Data Science and Artificial Intelligence. His expertise lies in the machine learning and deep learning arena. He is known to be a seasoned professional in the area of Business Consulting and Business Problem Solving, fuelled by his proficiency in machine learning and deep learning. He has been associated with organizations such as Mercedes-Benz Research and Development (India), Fidelity Investments, Royal Bank of Scotland among others. He has accumulated a diverse exposure through industries like BFSI, telecom and automobile. Rahul has also got papers published in IIM and IISc Journals.

Chiheb Chebbi is a Tunisian infosec enthusiast, author, and technical reviewer with experience in various aspects of information security, focusing on investigations into advanced cyber attacks and researching cyber espionage. His core interests lie in penetration testing, machine learning, and threat hunting. He has been included in many halls of fame. The proposals he has put forward with a view to giving presentations have been accepted by many world-class information security conferences.

DatTran is currently co-heading the data team at idealo.de, where he leads a team of data scientists and data engineers. His focus is to turn idealo into a machine learning powerhouse. His research interests range from traditional machine learning to deep learning. Previously, he worked for Pivotal Labs and Accenture. He is a regular public speaker and has presented at the PyData and Cloud Foundry summits. He also blogs about his work on Medium. His background is in operations research and econometrics. He received his MSc in Economics from Humboldt University, Berlin.

If you're interested in becoming an author for Packt, please visit authors.packtpub.com and apply today. We have worked with thousands of developers and tech professionals, just like you, to help them share their insight with the global tech community. You can make a general application, apply for a specific hot topic that we are recruiting an author for, or submit your own idea.

Machine learning involves developing and training models to predict future outcomes. This book is a practical guide to all the tips and tricks related to machine learning. It includes hands-on, easy-to-access techniques on topics such as model selection, performance tuning, training neural networks, time series analysis, and a lot more.

This book has been tailored toward readers who want to understand not only the concepts behind machine learning algorithms, but also the mathematics behind them. However, we have tried to strike a balance between these two.

If you're a machine learning practitioner, data scientist, machine learning developer, or engineer, this book will serve as a reference point for building machine learning solutions. You will also find this book useful if you're an intermediate machine learning developer or data scientist looking for a quick, handy reference to all the concepts of machine learning. You'll need some exposure to machine learning to get the best out of this book.

, Quantification of Learning, builds the foundation for later chapters. First, we are going to understand the meaning of a statistical model. We'll also discuss the thoughts of Leo Breiman about statistical modeling. Later, we will discuss curves and why they are so important. One of the typical ways to find out the association between variables and modeling is curve fitting, which is introduced in this chapter.

To build a model, one of the steps is to partition the data. We will discuss the reasoning behind this and examine an approach to carry it out. While we are building a model, more often that not it is not a smooth ride, and we run into several issues. We often encounter overfitting and underfitting, for several reasons. We need to understand why and learn how to overcome it. Also, we will be discussing how overfitting and underfitting are connected to bias and variance. This chapter will discuss these concepts with respect to neural networks. Regularization is one of the hyperparameters that is an integral part of the model building process. We will understand why it is required. Cross-validation, model selection, and 0.632+ bootstrap will be talked about in this chapter, as they help data scientists to fine-tune a model.