You will learn the steps required to build a successful machine learning application using Python and the scikit-learn library. Authors Andreas Müller and Sarah Guido focus on the practicalities of using machine learning algorithms and not the math behind them. If you are familiar with the NumPy and matplotlib libraries, you will find even better use for this book.
With this book you will learn:
- Basic concepts and applications of machine learning
- Advantages and disadvantages of widely used machine learning algorithms
- How to present the data processed by machine learning, including what aspects of the data to focus on
- Advanced methods for model evaluation and parameter adjustment
- The pipeline concept for chaining models and capsules of your workflow
- Methods for working with text data, including specific text processing techniques
- Tips for improving your machine learning and data science skills.
This is a very good book, and I'd say it's great for those unfamiliar with Python (of course, you won't be able to run the code). For anyone with a basic understanding of linear algebra / statistics, the authors can present all the important (and sometimes subtle but significant) details without the use of equations and most importantly their relationship to one another.
All of the concepts mentioned here are heavily aided by well thought out and well presented numbers, so again I would suggest that it is not even necessary for Python to understand. If you're familiar with Python, loading records and reproducing numbers are just a few lines of easy-to-understand code (with the exception of the mglearn include library which does a bit of "plot magic" for you. However, I believe in everyone .) di appropriate. You can ignore them and make the graphs your own way, or just print out the variables, it may not seem kind to post them).
I usually hesitate to buy books that claim to explain algorithms without equations, and I expect them to be cookbooks with techniques that are easily and incoherently explained (like an encyclopedia). I don't think this is the case for this book, however. The efficiency of scikit-learn is demonstrated and the algorithms behind it are intuitively explained and shown how they complement and complement each other.
As with any introductory reading, an addition is required from time to time, and the authors' reference to Statistical Learning Elements is helpful (difficult equation). There are places in the book where the author refers to statistical learning elements. I have found these points appropriate as further explanation would be unattainable.
I read this book in my spare time on vacation and most of the time I didn't have access to a computer. The concepts were presented so well that it was just fun to read. When I finally had time to log into a computer, I could try the techniques on my records by flipping the book back and forth, but otherwise with little effort.
Finally, being a researcher myself, I would recommend this book to any other researcher willing to delve into the world of machine learning. Further reading will always be required, but this book will give you an intuitive understanding and overview of the subject so that you will know what to do next and how to do it without going around in circles. Better yet, you've probably already applied it to your research!
Since I'm a seasoned Python person and have played with sklearn, this book has helped me turn a hobby into a functional product with paying customers. Andreas is not only an expert in the field and one of the leading developers of sklearn, but also has a great ability to explain concepts simply and intuitively. The reviews of the code examples lack the forest for the trees. It is really the simple but practical explanations of concepts like grid search implementation, pipelines, cross-validation, etc. that make this book valuable. While some of this information is available through the online documentation, Andreas's perspective on how to put it together is worth reading.
Highly recommended for anyone looking to get into applied machine learning or using sklearn.
I attended Andreas' ODCS sessions where he goes through examples and adds color comments. A clear writer / speaker - Very good, I look forward to your next books
I bought this book to give me a quick start on an independent study project "Introduction to Machine Learning". Of the books I bought for the same assignment, this was by far the most useful for creating hands-on machine learning applications.
The book is a great introduction to the Scikit-Learn framework, which I think is an extremely elegant machine learning toolkit.
Reading this book helped me improve the quality of the code I developed for the project, which greatly accelerated the speed at which I could get new results for the project.
If you're looking for a very theoretical machine learning text, you should look elsewhere. If you're looking for a guided introduction to the bread-and-butter tools of a great machine learning framework in Python, buy this one.