This is the third edition of the book, updated and seasoned, and my first time looking at it. Why learn and use Deep Learning? "DL techniques can solve problems with a level of accuracy that was not possible using previous methods."The book is nicely concise and thorough, well-written. Following the code example in the first chapter, I quickly fit the Sentiment Analysis model of IMDB reviews. I had not really used Google Colab before, it was easy and similar to Jupyter notebooks. You can choose to run on a CPU, GPU, or TPU. This first example uses the simplest of three methods of model building with tf.keras, the Sequential() model. Skimming the code made me curious - what is this and that?, so I searched online for the documentation, quickly found it at tensorflow dot org, where they also have tutorials. There are many code examples in the book and they use Python which uses "TensorFlow 2.x, a modular network library based on Keras-like APIs".I like the chapter divisions and the offerings; there are 20, which includes one focusing on the math behind DL. Other topics of interest to me are: Transformers, Probabilistic Tensorflow, Intro to AutoML, Four generations of TPUs, Other Useful DL libraries, ML Best Practices, and TensorFlow Lite. I like that there is a list of references and resources at the end of each chapter.I think this book will be an excellent companion on a further journey of exploration of DL model building. The library comes with datasets, if you want to avoid preparing your own at the start.

Absolutely amazing book which delivers insights on machine learning and NLP models. The mathematical and structural descriptions are well motivated and followed by code that is well documented using standard packages. It is rare to find such a reference on even one of the topics, but this reference delivers across a wide range of techniques.I was especially impressed with the chapters devoted to natural language processing. After well written chapters on basic concepts such as word vectors, the authors provide excellent coverage of transformers which are the current state of the art for language processing. The authors cover the basics of transformers and then illuminate the differences amongst the many transformer variates with their target uses and particular strengths. As in the other chapters, the discussion of transformers is capped by a detailed walk through of code insuring that the reader understands the steps needed to construct the processing pipeline through to model training and output.The ending chapters make up an excellent reference manual of concept and techniques such as parameter turning using AutoML, the mathematical methods used to optimize model coefficients by backpropagation, hardware decisions, and an introduction to other deep learning libraries.I highly recommend this book regardless of your level of modeling experience.Elliot NomaLead Data ScientistThe Financial Regulatory Authority

Really enjoyed reading through 'Deep Learning with TensorFlow and Keras'. The authors have delivered a comprehensive and detailed book on how to use TensorFlow and Keras. Not only will you get familiar with using ML platforms and open-source libraries, you will learn when and why you should use certain ML techniques. There is so much useful content here that I will plan to continue to use this book as a reference!

This textbook is really good, and contains from scratch knowledge about deep learning framework and implementation.Consider this textbook for the serious life long learners of deep learning, and also helpful in clearing the tensorflow developer exam.