What do you think?
Rate this book
Introduction to Machine Learning with Python: A Guide for Data Scientists
Machine learning has become an integral part of many commercial applications and research projects, but this field is not exclusive to large companies with extensive research teams. If you use Python, even as a beginner, this book will teach you practical ways to build your own machine learning solutions. With all the data available today, machine learning applications are limited only by your imagination.
You'll learn the steps necessary to create a successful machine-learning application with Python and the scikit-learn library. Authors Andreas Muller and Sarah Guido focus on the practical aspects of using machine learning algorithms, rather than the math behind them. Familiarity with the NumPy and matplotlib libraries will help you get even more from this book.
With this book, you'll learn:
Fundamental concepts and applications of machine learning
Advantages and shortcomings of widely used machine learning algorithms
How to represent data processed by machine learning, including which data aspects to focus on
Advanced methods for model evaluation and parameter tuning
The concept of pipelines for chaining models and encapsulating your workflow
Methods for working with text data, including text-specific processing techniques
Suggestions for improving your machine learning and data science skills
You'll learn the steps necessary to create a successful machine-learning application with Python and the scikit-learn library. Authors Andreas Muller and Sarah Guido focus on the practical aspects of using machine learning algorithms, rather than the math behind them. Familiarity with the NumPy and matplotlib libraries will help you get even more from this book.
With this book, you'll learn:
Fundamental concepts and applications of machine learning
Advantages and shortcomings of widely used machine learning algorithms
How to represent data processed by machine learning, including which data aspects to focus on
Advanced methods for model evaluation and parameter tuning
The concept of pipelines for chaining models and encapsulating your workflow
Methods for working with text data, including text-specific processing techniques
Suggestions for improving your machine learning and data science skills
398 pages, Paperback
First published June 25, 2015
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 51 reviews
May 20, 2018
I do not like this book as much as as An Introduction to Statistical Learning: With Applications in R, but if you are constrained or committed to using Python instead of R, it is the best available alternative as of 2018, and I do think it's pretty good overall. You will need to already be fluent in numpy, pandas, and matplotlib to read this book; if you area not, read Python Data Science Handbook: Tools and Techniques for Developers first.
December 9, 2016
What a useful book - it focuses mostly on scikit-learn with some numpy, pandas and matplotlib thrown in, you could say it's an in-depth tour of some of the more useful methods in scikit-learn - classifying, regression, a bit of clustering, PCA, all the different ways to measure the outcome of your model, how to use the incredibly useful scikit-learn Pipeline to test parameters and models, etc.
The examples are useful and interesting, especially the face picture clustering and classification is amazing. There's not much mathematics involved, if I remember correctly there's maybe two formulas in the whole book. However, parameters play a large role - if there's a new method presented there's usually a bit of discussion which parameters have the biggest impact on the outcome. There's also quite a bit of wisdom on the many dangers of overfitting and choosing the wrong test and training set (or in the case of cross-validation, the danger of training too much towards the test set).
If you're looking for Artificial Neural Networks then this may be a good start, the chapter on ANNs has a very intuitive explanation of the logic behind layers and neurons, but the focus of the book is elsewhere, which the authors explicitly state (in the last chapter the authors recommend Goodfellow et al.'s Deep Learning for that - $70 and no epub/mobi, ouch).
One minor thing I'm not sure about: to skip wasting paper on fiddling with matplotlib the authors often call their own mglearn library for plots and some utility functions, I'm not sure whether that 'hides' too much complexity from the reader - but again, from what I'm seeing it's mostly matplotlib fiddling that's hidden in those functions, which is enormously frustrating and has no place in this book.
Recommended for: semi-advanced Python programmers who'd like to learn new ways to make new connections from their data
Not recommended for: people who want to know what all that artificial neural network hype is about.
Maybe: People who've already devoured the amazing scikit-learn documentation, some of the examples in the book are straight from the documentation (clustering), the more complex examples are 'novel'
The examples are useful and interesting, especially the face picture clustering and classification is amazing. There's not much mathematics involved, if I remember correctly there's maybe two formulas in the whole book. However, parameters play a large role - if there's a new method presented there's usually a bit of discussion which parameters have the biggest impact on the outcome. There's also quite a bit of wisdom on the many dangers of overfitting and choosing the wrong test and training set (or in the case of cross-validation, the danger of training too much towards the test set).
If you're looking for Artificial Neural Networks then this may be a good start, the chapter on ANNs has a very intuitive explanation of the logic behind layers and neurons, but the focus of the book is elsewhere, which the authors explicitly state (in the last chapter the authors recommend Goodfellow et al.'s Deep Learning for that - $70 and no epub/mobi, ouch).
One minor thing I'm not sure about: to skip wasting paper on fiddling with matplotlib the authors often call their own mglearn library for plots and some utility functions, I'm not sure whether that 'hides' too much complexity from the reader - but again, from what I'm seeing it's mostly matplotlib fiddling that's hidden in those functions, which is enormously frustrating and has no place in this book.
Recommended for: semi-advanced Python programmers who'd like to learn new ways to make new connections from their data
Not recommended for: people who want to know what all that artificial neural network hype is about.
Maybe: People who've already devoured the amazing scikit-learn documentation, some of the examples in the book are straight from the documentation (clustering), the more complex examples are 'novel'
May 29, 2018
I have not read much of this book, to be honest; my teacher covered all literature during his classes so I didn't feel a need to read what I had heard and seen already. But this book is really nicely written: it patiently teaches all you need to know when you start with little to no foreknowledge, providing codes and its outputs and explaining what this output means.
I would recommend this to anyone who wants to learn how to do machine learning; familiarity with Python is a big pre.
I would recommend this to anyone who wants to learn how to do machine learning; familiarity with Python is a big pre.
October 20, 2018
This is a great book.
It is a nice introduction to Machine Learning (scikit-learn specifically) without much maths needed. It will by no means make you an expert, but it will give you a good sense of the basics, a walkthrough of scikit-learn and hopefully some intuition about the popular algorithms. A nice supplement to it is the Andrew Ng Coursera course about Machine Learning. The two make a very good starting point into the ML journey for programmers who are curious.
It is a nice introduction to Machine Learning (scikit-learn specifically) without much maths needed. It will by no means make you an expert, but it will give you a good sense of the basics, a walkthrough of scikit-learn and hopefully some intuition about the popular algorithms. A nice supplement to it is the Andrew Ng Coursera course about Machine Learning. The two make a very good starting point into the ML journey for programmers who are curious.
November 20, 2018
This book is very basic introduction to Machine Learning and there are better books for example hands on machine learning with scikit-learn and tensorflow. The examples in the book uses a library that the author did which makes difficult to really learn how to do the analysis in python
Read
June 21, 2019finally, I could read this great book. I have found that some topic that I didn't know about them like chaining and pipeline. the working with text data has more profit for me
July 26, 2020
Ótimo livro, realmente cumpre seu propósito de ensinar o básico para criar aplicações de machine learning com python. Recomendo muito para pessoas que tenham interesse no tópico e procuram por onde começar.
September 3, 2019
It might be my bad choice, but this book doesn't give you a nice insight into how (classical) machine learning algorithms and techniques work. It is more or less a restructuring of scikit-learn's documentation with nearly zero mention of mathematical backgrounds of the techniques used. It is full of code examples and the results of running them and an explanation of "how" those results are and often not "why". The most useful part of the book is the last chapter which refers the reader to papers and books better explaining the "why"s and less "technical" details. That chapter is not well-edited and contains a handful of typos. (Anyway it is not a big deal)
January 24, 2020
ML es la versión marketinera de la econometría, con la posibilidad de aplicarlo a mil cosas distintas. El libro intenta hacer una introducción a todo eso, con lo cual termina tocando temas difícil, rápido y con poca aplicación práctica real. Pero es al menos una introducción... Da para seguir leyendo por esta línea...
April 1, 2020
If you are new to the topic of Machine Learning, this is definitely the go-to book. While staying away from detailed mathematics, this book gives a good overview of the most common techniques used in the field. It was everything that I expected and more. Basic knowledge of python programming is needed.
January 8, 2018
It is a nice book to start Machine Learning. Book explains about all the algorithms.
Read
February 16, 2022Don't have the audacity to rate a book like this...yet.
July 17, 2019
This book should be the first book for anyone who has a bit of programming background and want to overview how machine learning would look like without deep diving into the linear algebra and/or any relevant math.
The book uses Python, scikit-learn, bumpy, etc that are well defined and have been widely used, and take examples one by one, but not with serious math or from the scratch but using existing scikit-learn.
Probably some people would like to learn all the stuff from the scratch, including how it works and what's the math behind it, etc etc, but there are other people like me who what to have brief overview of what's it look like and how it's going to work first, and then gradually dive into inch by inch.
This book will give you great overview of what noways machine learning look like, and also give you know those well defined APIs/library with consistency could help library users.
Highly recommend this book.
The book uses Python, scikit-learn, bumpy, etc that are well defined and have been widely used, and take examples one by one, but not with serious math or from the scratch but using existing scikit-learn.
Probably some people would like to learn all the stuff from the scratch, including how it works and what's the math behind it, etc etc, but there are other people like me who what to have brief overview of what's it look like and how it's going to work first, and then gradually dive into inch by inch.
This book will give you great overview of what noways machine learning look like, and also give you know those well defined APIs/library with consistency could help library users.
Highly recommend this book.
November 28, 2018
Great way to get started with Python and ML:
- Gives overview of tools/libraries you'll likely need
- Broad overview of algorithms, with a good explanation on how they work and insight into how the main parameters influence behavior (with examples in the book and code to demonstrate how to use. Code is also available on Github.)
- Many guidances as when to use what (depending on which kind of data/problem you've got; which techniques work better + it gives an idea on which techniques are typically used, and which only in specific cases + informs on trade-offs between model accuracy and model complexity)
The chapter on text data is fairly short (more like a managers overview). However, at this stage of learning, it's sufficient for me. It does provide references to other books when you want to go into specific details.
- Gives overview of tools/libraries you'll likely need
- Broad overview of algorithms, with a good explanation on how they work and insight into how the main parameters influence behavior (with examples in the book and code to demonstrate how to use. Code is also available on Github.)
- Many guidances as when to use what (depending on which kind of data/problem you've got; which techniques work better + it gives an idea on which techniques are typically used, and which only in specific cases + informs on trade-offs between model accuracy and model complexity)
The chapter on text data is fairly short (more like a managers overview). However, at this stage of learning, it's sufficient for me. It does provide references to other books when you want to go into specific details.
December 11, 2018
I would recommend all pure beginner to start machine learning from this book, Author Andreas clarifies several key concepts (for example grid search and cross validation) in details, which confuses me for long, now I am so clear. Another good point is that the book is so well structured, and could read a second time, I recognise that I couldn't find anther more efficient path!
Meanwhile, I found a code error through all the code in this book, in Chapter 5, Evaluation Metrics and Scoring (page 286), I got "Unsupported operand type(s) for //: 'tuple' and 'int'", probably the newer version Python 3 causes this error, only influence one cell input, nothing serious.
All in all, I am highest recommend this book, and thank you Andreas
Meanwhile, I found a code error through all the code in this book, in Chapter 5, Evaluation Metrics and Scoring (page 286), I got "Unsupported operand type(s) for //: 'tuple' and 'int'", probably the newer version Python 3 causes this error, only influence one cell input, nothing serious.
All in all, I am highest recommend this book, and thank you Andreas
August 7, 2020
If you want to understand Machine Learning, start here. The facial recognition sections alone are worth the price of admission. Deep enough to be interesting and informative without overwhelming the reader. Take it slow and read it with a pencil and a computer nearby (the examples are on GitHub).
October 17, 2020
Coming into machine learning with no computer science or math background, Muller’s textbook provides a great introduction to the field of data science. The textbook covers many of the important machine learning algorithms used by practitioners, explaining how the algorithms make predictions and some of the strengths/weaknesses to consider. This isn’t the right book to learn the mathematics (Introduction to Statistical Learning is better for that), but it’s a great first book because it won’t spook you away from the field. I think good follow up would be Hands-on Machine Learning and ISLR or Elements of Statistical Learning. I think I spent about a week and a half to two weeks going through the book for 2-4 hours a day.
November 9, 2022
This is a very good book to start with basics of Machine learning.
It will assume you have some basic to middle level proficiency in Python. It will also be useful if you have some background on statistics (or my case Econometrics). This will accelerate the learning process.
At the end of the day, considering the scope of the book, it will provide you with a basic guide in how to apply and evaluate most common ML techniques, assuming other problems (including data assessment) are already taken care of.
That being said. After reading you will not be an expert on the subject, but at least you will know what is out there.
It will assume you have some basic to middle level proficiency in Python. It will also be useful if you have some background on statistics (or my case Econometrics). This will accelerate the learning process.
At the end of the day, considering the scope of the book, it will provide you with a basic guide in how to apply and evaluate most common ML techniques, assuming other problems (including data assessment) are already taken care of.
That being said. After reading you will not be an expert on the subject, but at least you will know what is out there.
December 15, 2020
This is a very good book for understanding the basics of machine learning. It walks readers through each algorithm, explaining what they are doing, how they work, and when to use them. The downside of this book is that the Python examples rely on a package written specifically for the examples in the book. The code does not use the traditional libraries that data scientists would typically use.
August 17, 2023
A great book giving the overview of machine learning. This book not only explains the machine learning, but provides the pointers for next level of understanding.
Though some example codes in this book are not outdated and cannot work as they are, most of them can be fixed by renaming of methods or argument names.
Though some example codes in this book are not outdated and cannot work as they are, most of them can be fixed by renaming of methods or argument names.
September 3, 2023
Good book to start and get to know some important things about machine learning, but it is also missing other important topics, such as machine learning for Time Series Forecasting (predict future sales). All in all, a decent read and the first book review I am sharing as DATASOFI. Check datasofi.tech to get in touch.
May 5, 2020
Great and easy to read introduction of machine learning using Python. If you are looking for a book that explains the basic machine learning workflow in layman’s terms and sans lots of theory this book is for you.
June 24, 2020
The best book for beginner in Machine Learning thus far. All other books I had read paid too much attention to maths which made me lost (because I am no good at it). But, this book is different. The author pays attention to the concepts and puts the difficult vocabs inside parentheses.
August 2, 2017
Definitely introductory, but also quite thorough. Well organized as well.
November 22, 2017
Very good summary about scikit-learn library practical usage.
August 19, 2018
Interesting book, it provides a good introduction to the applications of machine learning, without spending too much time on the mathematics.
September 1, 2018
Nice intro to machine learning in python. The book is easy and intuitive for beginner machine learning practitioners.
September 7, 2018
Focused on statistical learning, but cover very good fundamental, like model search and all that.
December 26, 2018
fine book for starter
June 20, 2019
ML basis book.
Displaying 1 - 30 of 51 reviews