Machine learning and deep learning are two research areas that have created excitement and interest in many application domains. By using Python as the implementation medium, and by releasing most of the initial example set as an accompanying resource with this book, John Hearty appears to be determined not to leave out any interested person from participating in the development of intelligent solutions.

Advanced machine learning with Python is organized in nine chapters: “Unsupervised Machine Learning,” “Deep Belief Networks,” “Stacked Denoising Autoencoders,” “Convolutional Neural Networks,” “Semi-Supervised Learning,” “Text Feature Engineering,” “Feature Engineering Part II,” “Ensemble Methods,” and “Additional Python Machine Learning Tools.”

A notable feature that strikes anyone reading this book is the richness of examples provided along with the datasets, as part of accompanying resources downloadable from the publisher’s website. The code is designed to work with Python version 3 and makes use of the libraries associated with machine learning like sklearn, theano, and tensorflow, in addition to the libraries numpy, scipy, matplotlib, urllib, and xgboost.

The examples are also handpicked to expose interested developers to the domains of intelligent solutions like natural language processing, image processing, data mining, and signal and speech processing. These make the book unique.

Hearty has made the best use of his experience in developing solutions for industry, and this book, according to his own words, is a “mixture of toy examples, dissection of expert approaches, and more real-world challenges.”

In my opinion, this book would be of great interest and use to budding researchers keen to pursue their skills in advanced machine learning and deep learning, and more specifically to those who are looking to migrate from a proprietary MATLAB environment to a free and open-source Python environment.

More reviews about this item: Amazon, Goodreads