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

I've been following this book since its first edition, about time I write a review! It really does strike the perfect balance between code and theory. Everything is clear and written in a friendly tone. It'll get you started in applying everything from basic linear regression through decision tree, all the way to deep learning. My favorite is chapter 2, which is a step-by-step guide on exploring a data project, it's like having a professional guide you. I'm an experienced software developer, and I owe this book a lot for introducing me to many concepts. I'm old-school, so sitting down with a book and copying code examples takes me back and is a familiar experience. For some people, copy pasting might be more intuitive but you really can learn from doing things by hand. The full code is on github, but I recommend using it for reference only. What this book isn't, and doesn't pretend to be, is an introduction to Python. Some basic programming knowledge is needed, but if you want to work in the field, you'd need that anyway, and you shouldn't be afraid to dive into it. Looks like I'll be checking the 3rd edition!

I'm currently getting my MS in health data science and this was the book we had to get for my machine learning class. I was annoyed when the teacher said the class would be textbook heavy and he was only going lecture on high level concepts, I thought there was no way textbook would be able to a carry a class and boy was I wrong. This is hands down the best textbook I've ever bought! I never expected a data science text book to be easy to read but this book flows so well!, its easily digestible and it gives great examples with data that is easily available. You can write completely functional ML code from this book alone but one of the best features is that the book has GitHub site broken down chapter by chapter that helps fill the code out. If you are someone like me who hadn't had any experience with Matplotlib the github was super helpful because it covers in depth how to make really nice plots for the various models. I would recommend this book to anyone who is doing machine learning. The only thing I would change about this book is when it gets into decision trees, RF, various boosting types, XGB, as it moves through the models it only gives an example of the classification form of the model or the regression for of the model and I think it would be helpful if it gave examples for both for each model. But with that being said this was a pretty minimal thing I would change and I would still buy the book again even if they didn't change it! It's definitely worth the money!

The Tokyo Olympics of 2020 got postponed to 2021. If there were a contest for best AI/ML book at the Olympics this year this book would have earned the gold medal ! I loved it so much that I read it at least twice, and each time I underlined/highlighted/took-notes. I love how lucidly the author explains concepts. He does an excellent job of explaining topics such as the model, the learning algorithm (also called the optimization algorithm), regularization hyperparameter, generalization etc. The examples are great and even if one does not know python programming it is easy to follow along. (I learned python a few months later, which made it even easier and more interesting to follow the examples in this and other books). While no one single book can teach one ML/AI, this book would make the Mount Rushmore of AI/ML books (along with (1) Intro to Statistical Learning by Hastie etc (2) Intro to Machine Learning by Alpaydin (3) Deep Learning by Goodfellow, Bengio etc). I highly recommend this book to anyone aspiring to get into the field of ML/AI.

I took a machine learning graduate course in my master program. I had a top conference paper. The professor used 1st edition of this book as one textbook for the course. I had a 1st edition of the book but did not have time to read. Now I buy the 2nd edition because the Tensorflow 2 has merged with Keras, which means we can avoid to learn the hard syntax of tensorflow 1.0, and there are a lot of new advances in machine learning, such as generative models. Also to my surprise, the book is colorful. That makes the book is more interesting. Each chapter has summary of math. That is better than some programming machine learning books that do not have any math. If you have some backgrounds in math of machine learning, this book can save you time because it gives you the whole picture without lost. If you are very interested in some equations and want to derive them, you can use Pattern Recognition and Machine Learning book. The Github has a lot of python projects of machine learning. The codes are well-written. If you can write codes like the codes in the projects, you will have the potential to enter Google. Go Google, the book is a must have.

This book covers many topics of ML and explains them with good examples. However, I believe it should be a little bit tough for a beginner. Similarly, it could not be the best book for an advanced reader because it gives pointers for advanced topics but does not go in-depth like mathematical explanation. In summary, it is an excellent book if you are looking for real-life examples with python code and you have a good basic idea in ML.

Livro excelente e muito bem didático.

This book should be regarded as a "gold-standard" for technical books. It balances theory and practice, has exercises (actually with answers!) and covers a tremendous breadth and depth. The book starts out in a refreshingly unconventional way of giving you a crash course in ML concepts before diving in to an end-to-end project. I note that one reviewer didn't like that but I liked it a lot. While a lot of it will go over your head if you lack experience (and the author assumes you don't have much), it gives you appreciation of what an overall real-life project might look like. The rest of the book is spent unpacking each of those stages. The first part of the book looks at more "classical" or traditional machine learning concepts like linear regression, logistic regression, SVMs, decision trees, ensemble learning and unsupervised models. Along the way you learn a lot of data science best-practises and how to train and test things properly. The second part dives into deep learning, progressing from general neural networks to CNNs, RNNs, LSTMs, autoencoders and GANs. You get a flavour of how GPT models work. Other topics covered in this section are Tensorflow and Keras (including a part on deploying models) and a chapter on another paradigm: reinforcement learning. Geron doesn't shy away from the math but gives you enough theory to appreciate the detail if you like that, and explains it in intuitive ways and with code. Some of the formulas can look intimidating but they are unpacked and explained well. There are review questions and/or exercises at the end of each chapter. One of my biggest frustrations with technical books in general is when they give you questions but no answers. Here, you get answers and also worked code in the provided notebooks, which is amazing. Other technical authors: take note. The exercises are often quite challenging to implement or at least open-ended, but I believe that to be a good thing. I learnt a lot from doing them (I'll admit I didn't do all of them!). The writing is clear, engaging and often humourous. To sum up, if you want to learn more about ML, I highly recommend this book. This review is for the 2nd edition but I'll be buying the 3rd edition and will definitely be re-reading. There is so much great information to take in. Thanks to the author for this masterpiece.

Excellent book for getting into machine learning. Plenty of example code.

Good Packaging done. Great Job.

This second edition book is totally worth your money