The Deep Learning with Keras Workshop: An Interactive Approach to Understanding Deep Learning with Keras, 2nd Edition

by Matthew Moocarme

The Deep Learning with Keras Workshop

446 pages, Paperback

Published February 28, 2020

Reviews

[Marcos Malumbres · Rating 5 out of 5]

Very good introduction to the scikit-learn, tensorflow and keras frameworl, including basics of machine learning with sklearn, basic neural networks, techniques for model selection and improving efficiency. The book includes special chapters for Convolutional Neural Networks and image analysis and a brief introduction to Recurrent Neural Networks.

Overall, it is an excellent introduction to all these topics, with a basic explanation of the concepts and implementation without getting deep into the mathematics. The exercises are quite good, a bit repetitive in some cases but all ok! I found (pages refer to the 2nd edition) one mistake in adding a parameter to a function (page 187), one .model.fit_generator() deprecated command (page 239), an internal hidden layer with input_shape (page 245), a typo (datasetset) in page 272, and an unexpected problem with the dimensions of one array (page 296). Whereas I easily solved the first ones, I was unable to fix the last problem. In any case, very few errors for a 430-page book full of codes.

Highly recommended for intro into Keras!!!