Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems 1st Edition

Hands-On Machine Learning strikes a perfect blend between application and theory. Beginners to machine learning will find it clear to follow and will be able to build complete systems within a few chapters while those with an intermediate level of experience will find a comprehensive, up-to-date guide to this exciting field.Pros:+ Practical: The book focuses on examples and implementations of the algorithms rather than the mathematics allowing readers to quickly build their own machine learning models+ Readable: Geron does not get too caught up in the details, and he provides warnings when the next section is heavy on theory+ Online Jupyter Notebooks: The Jupyter Notebooks that accompany this book (and can even be viewed for free with no purchase from the author's GitHub) are worth the entire purchase price. They feature examples of all the code in the book, plus additional explanatory material. The end-of-chapter solutions to the coding exercises are gradually being added to the notebooks.+ Up-to-date: The leading edge of machine learning (and in particular deep learning) is constantly shifting, and Geron does his best to keep the notebooks updated. Multiple times I have read an ML paper and then found the technique implemented in the notebooks within weeks of the publication of the article. Some of the techniques in the book may not be at the absolute forefront of the field, but they are still good enough for learning the fundamentals.+ Engaging: The book is a joy to read, and the author is quick to respond to issues pointed out by readers in the book or in the Jupyter Notebooks. Clearly, the author enjoys machine learning and teaching it to others.Cons:- Experts may find this book lacks enough depth because it is more focused on getting up and running rather than optimization. It also is specifically aimed towards Python (and Tensorflow for deep learning) so those looking for implementations in other frameworks will have to search elsewhere.- Due to the rapidly-evolving nature of the field, a print book on machine learning will always need to be periodically re-issued to stay on top of all the developments. Nonetheless, the fundamentals covered in this book will remain relevant and the Jupyter Notebooks are constantly updated with new techniques.Final Line: If you have some basic experience with Python (loops, conditionals, dictionaries, and especially Numpy) and zero to a medium level of experience with machine learning, this book is an optimal choice. I would recommend it both for those wishing to self-study and quickly develop working models, and for students in machine learning who want to learn the implementations of more theoretical coursework. I have enjoyed spending time working through the chapters and the exercises and have found this book extremely useful.

This has to be at the top of my list of most highly recommended books! The amount of material it covers is awesome, and I can find almost no fault with it. The writing is extremely clear, easy to read, written in impeccable English. Very well edited. I don't think I came across any spelling or grammar errors, or any real errors at all. Truly solid writing.The breadth of information covered if quite wide. The choice to start with Scikit-Learn was interesting, but makes sense on some level while he's introducing the more basic machine learning concepts. Simple machine learning techniques like logistic regression, data conditioning, dealing with training, validation, test set. Even if you've read about these concepts a million times, you might still glean useful information from these pages.The Tensorflow section is also super well done. Straightforward setup instructions, pretty intelligible explanation of the basic concepts (variables, placeholders, layers, etc.) to get you started. The example code is quite good, and the notebooks are quite complete and seem to work well, with maybe a few tweaks and additional setup for some. I also found that the notebooks show more examples than what's in the book, which can be nice.I only went really hands on with the reinforcement learning notebook, and found that it was well done and a good base to start my own work from. Even just having a section on reinforcement learning is very rare in a book of this style, and Geron's samples and explanations are really solid. He obviously has a strong grasp of many varied fields within deep learning, and that includes reinforcement learning. The only thing I wish it had was an A3C sample, to make my life that much easier. But you can't have everything.I really liked his tips on which types of layers, activations, regularization, etc. are most effective, and gives good starting points for decent convergence. His explanation of multi-GPU Tensorflow was also quite good. The Tensorboard section was also very useful.In short, if you want ONE book to get you into machine learning, and Tensforlow is on your radar, you can't go wrong with this one. Highly recommended!

This book has been the greatest resource I've found to learn machine learning. I have intermediate experience with Python and had only a basic understanding of Jupyter notebooks and scientific Python before opening this book, and it's been an awesome experience.This book is:- clearly written- teaches you machine learning through projects rather than jargon- contains a lot of personality (As quoted in a note, "You will often see people set the random seed to 42. This number has no special property, other than to be The Answer to the Ultimate Question of Life, the Universe, and Everything.")- contains exercises at the end of each chapter to recap what you learnedThis is by no means an exhaustive resource to scientific Python or machine learning, but I've obtained a lot of insight into the methodologies of a data scientist, as well as seeing Pandas methods put into action. Aurélien will provide you a basic understanding of some of the things he does, but it is up to you to use the docs if you want to see the power of some of the methods. With all the things you can do in matplotlib, pandas, numpy, and sci-kit learn, I appreciate seeing them in action, then doing research on them myself rather than sifting through docs to see what methods, functions, or attributes I can use.I have yet to reach the section on TensorFlow, but I'll give an update when I get there. Also, don't be intimidated by the length of this book, it's NOT a textbook.