Books are not the same especially on Deep Learning
It feels and really is that we are bombarded by a growing number of books on Machine Learning, especially Deep Learning. Due to this large number of books published it is quite difficult to tell what book is worth investing the time and effort. Since humans as spices have a constrained lifespan what books you choose to read matters. That is why you’d better read the best books available out there on the subject.
The paragraph above may sound a little bit as advertising, but I really think a good book, which is frankly a subjective definition makes a difference. I would say, that a good book in my opinion is the one that engages you, makes you think, at least a little bit, and what is important makes you strive to check the references it provides and find additional information beside what the book already includes.
The newest Deep Learning for Computer Vision book from Machine Learning Mastery brings exactly this. It is crafted in a well recognizable machinelearningmastery style which is a practical approach with a simple to complex information presentation spiced with just enough theory to get you started in the Machine and Deep Learning fields.
More details on content
- If you read any of their books previously you know that each chapter has a battle proven working Python code samples that work on MacOS, Linux, and even Windows (who would thought).
- Each chapter composed in such a way that it may serve as a stand alone tutorial, but overall they are tide together in a logical order if you prefer to read the book from beginning to end.
- What I personally find valuable is the Extensions at the end of the chapters that provide additional tasks to practice the chapter’s material.
- Not to mention the references to books, papers and other relevant data that were mentioned in the chapter.
More Technical information
The book’s subject is about Pytohn libraries to process images while working on machine learning or Deep Learning projects. The main library that is used for Deep Learning is Keras including helpful Keras Functional API. In addition, the book describes how to download build, train and run models such as Mask R-CNN, Multi-task Cascade CNN, FaceNet and other using TensorFlow.
If this post made you curious about the book then give it a try. You may find it very helpful.