Machine Learning Automation with TPOT

Dario Radecic

• Editorial: Packt Publishing
• Año: 2021
• ISBN:
  9781800564961
• Edición: 1
• Categoría: Ciencia de la computación
• Categoría: Tratamiento de datos

Discover how TPOT can be used to handle automation in machine learning and explore the different types of tasks that TPOT can automateKey Features• Understand parallelism and how to achieve it in Python.• Learn how to use neurons, layers, and activation functions and structure an artificial neural network.• Tune TPOT models to ensure optimum performance on previously unseen data.Book DescriptionThe automation of machine learning tasks allows developers more time to focus on the usability and reactivity of the software powered by machine learning models. TPOT is a Python automated machine learning tool used for optimizing machine learning pipelines using genetic programming. Automating machine learning with TPOT enables individuals and companies to develop production-ready machine learning models cheaper and faster than with traditional methods. With this practical guide to AutoML, developers working with Python on machine learning tasks will be able to put their knowledge to work and become productive quickly. You'll adopt a hands-on approach to learning the implementation of AutoML and associated methodologies. Complete with step-by-step explanations of essential concepts, practical examples, and self-assessment questions, this book will show you how to build automated classification and regression models and compare their performance to custom-built models. As you advance, you'll also develop state-of-the-art models using only a couple of lines of code and see how those models outperform all of your previous models on the same datasets. By the end of this book, you'll have gained the confidence to implement AutoML techniques in your organization on a production level. What you will learn• Get to grips with building automated machine learning models• Build classification and regression models with impressive accuracy in a short time• Develop neural network classifiers with AutoML techniques• Compare AutoML models with traditional, manually developed models on the same datasets• Create robust, production-ready models• Evaluate automated classification models based on metrics such as accuracy, recall, precision, and f1-score• Get hands-on with deployment using Flask-RESTful on localhostWho this book is forData scientists, data analysts, and software developers who are new to machine learning and want to use it in their applications will find this book useful. This book is also for business users looking to automate business tasks with machine learning. Working knowledge of the Python programming language and beginner-level understanding of machine learning are necessary to get started.

Section 1: Introducing Machine Learning and the Idea of Automation

This section provides a quick revision of machine learning – classification and regression tasks, an overview of automation and why it is needed, and what options are available in the Python ecosystem.

This section contains the following chapter:

• Chapter 1, Machine Learning and the Idea of Automation

Chapter 1: Machine Learning and the Idea of Automation

In this chapter, we'll make a quick revision of the essential machine learning topics. Topics such as supervised machine learning are covered, alongside the basic concepts of regression and classification.

We will understand why machine learning is essential for success in the 21st century from various perspectives: those of students, professionals, and business users, and we will discuss the different types of problems machine learning can solve.

Further, we will introduce the concept of automation and understand how it applies to machine learning tasks. We will go over automation options in the Python ecosystem and compare their pros and cons. We will briefly introduce the TPOT library, and discuss its role in the modern-day automation of machine learning.

This chapter will cover the following topics:

• Reviewing the history of machine learning
• Reviewing automation
• Applying automation to machine learning
• Automation options for Python

Technical requirements

To complete this chapter, you only need Python installed, alongside the basic data processing and machine learning libraries, such as numpy, pandas, matplotlib, and scikit-learn. You'll learn how to install and configure these in a virtual environment in Chapter 2, Deep Dive into TPOT, but let's keep this one easy. These libraries come preinstalled with any Anaconda distribution, so you shouldn't have to worry about it. If you are using raw Python instead of Anaconda, executing this line from the Terminal will install everything needed:

> pip install numpy pandas matplotlib scikit-learn

Keep in mind it's always a good practice to install libraries in a virtual environment, and you'll learn how to do that shortly.

The code for this chapter can be downloaded here:

https://github.com/PacktPublishing/Machine-Learning-Automation-with-TPOT/tree/main/Chapter01

Reviewing the history of machine learning

Just over 25 years ago (1994), a question was asked in an episode of The Today Show – "What is the internet, anyway?" It's hard to imagine that a couple of decades ago, the general population had difficulty defining what the internet is and how it works. Little did they know that we would have intelligent systems managing themselves only a quarter of a century later, available to the masses.

The concept of machine learning was introduced much earlier in 1949 by Donald Hebb. He presented theories on neuron excitement and communication between neurons (A Brief History of Machine Learning – DATAVERSITY, Foote, K., March 26, 2019). He was the first to introduce the concept of artificial neurons, their activation, and their relationships through weights.

In the 1950s, Arthur Samuel developed a computer program for playing checkers. The memory was quite limited at that time, so he designed a scoring function that attempted to measure every player's probability of winning based on the positions on the board. The program chose its next move using a MinMax strategy, which eventually evolved into the MinMax algorithm (A Brief History of Machine Learning – DATAVERSITY, Foote, K., March 26; 2019). Samuel was also the first one to come up with the term machine learning.

Frank Rosenblatt decided to combine Hebb's artificial brain cell model with the work of Arthur Samuel to create a perceptron. In 1957, a perceptron was planned as a machine, which led to building a Mark 1 perceptron machine, designed for image classification.

The idea seemed promising, to say at least, but the machine couldn't recognize useful visual patterns, which caused a stall in further research – this period is known as the first AI winter. There wasn't much going on with the perceptron and neural network models until the 1990s.

The preceding couple of paragraphs tell us more than enough about the state of machine learning and deep learning at the end of the 20th century. Groups of individuals were making tremendous progress with neural networks, while the general population had difficulty understanding even what the internet is.

To make machine learning useful in the real world, scientists and researchers required two things:

• Data
• Computing power

The first was rapidly becoming more available due to the rise of the internet. The second was slowly moving into a phase of exponential growth – both in CPU performance and storage capacity.

Still, the state of machine learning in the late 1990s and early 2000s was nowhere near where it is today. Today's hardware has led to a significant increase in the use of machine-learning-powered systems in production applications. It is difficult to imagine a world where Netflix doesn't recommend movies, or Google doesn't automatically filter spam from regular email.

But, what is machine learning, anyway?

What is machine learning?

There are a lot of definitions of machine learning out there, some more and some less formal. Here are a couple worth mentioning:

• Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed (What is Machine Learning? A Definition – Expert System, Expert System Team; May 6, 2020).
• Machine learning is the concept that a computer program can learn and adapt to new data without human intervention (Machine Learning – Investopedia, Frankenfield, J.; August 31, 2020).
• Machine learning is a field of computer science that aims to teach computers how to learn and act without being explicitly programmed (Machine Learning – DeepAI, Deep AI Team; May 17, 2020).

Even though these definitions are expressed differently, they convey the same information. Machine learning aims to develop a system or an algorithm capable of learning from data without human intervention.

The goal of a data scientist isn't to instruct the algorithm on how to learn, but rather to provide an adequately sized and prepared dataset to the algorithm and briefly specify the...

Índice

• Machine Learning Automation with TPOT
• Contributors
• Preface
• Section 1: Introducing Machine Learning and the Idea of Automation
• Chapter 1: Machine Learning and the Idea of Automation
• Section 2: TPOT – Practical Classification and Regression
• Chapter 2: Deep Dive into TPOT
• Chapter 3: Exploring Regression with TPOT
• Chapter 4: Exploring Classification with TPOT
• Chapter 5: Parallel Training with TPOT and Dask
• Section 3: Advanced Examples and Neural Networks in TPOT
• Chapter 6: Getting Started with Deep Learning: Crash Course in Neural Networks
• Chapter 7: Neural Network Classifier with TPOT
• Chapter 8: TPOT Model Deployment
• Chapter 9: Using the Deployed TPOT Model in Production
• Other Books You May Enjoy