Deep Learning as a subject is very theoretical with its foundation based on a wide variety of Mathematical tools and techniques. Without the proper understanding of these Mathematical tools, it is very difficult to fine tune and enginner a model to solve a real problem. So, it is very important for engineers working in ML or Deep Learning to understand this mathematical tools and their proper utility to solve a problem. On the other hand, general students or practitioners of Math often fail to translate their knowledge of Math into programming problems in right application areas.
Krishnendu Chaudhury with Ananya Ashok, Sujay Narumanchi and Devashish Shankar in the Manning Early Access Program (MEAP) for Math and Architectures of Deep Learning has done an excellent job in bridging this disconnect. This is not a book to learn Math Concepts, but it is certainly a book to understand the Architectures of Deep Learning and the role math techniques and tools play to solve real life Machine Learning problems. Python programmers working in real life machine or deep learning problems will be greatly benefited by assimilating this book. Sufficient Python code snippets have been provided throughout the book for almost all the concepts to help enable a reader famaliarize with the underlying math priniciples and how to code them in Python in platforms like Tensorflow and PyTorch. Unlike any other book, each and every line of code is associated with an explanation of why that piece of code is required.
The book starts with a beautiful overview of machine learning and deep learning with a toy example bundled with sufficient illustrations. Subesequent chapters have introduced different mathematical notions from the point of view of its applicability in machine learning. Each chapter clearly introduces the machine learning or data science problem and then has succiently showed how the problem can be formalized using mathematical tools and then have explained the solution. At the end of each topic or chapter, how the solution to the machine learning problem can be programmed using python is shown and that is a one of the biggest advantge to the readers. It not only just helps the reader to understand the concept, but readily can see the concept in action though implementing the solution in Python. Each concept in each chapter has been described with sufficient rigor and detail for the reader to understand the concept thoroughly. In our opinion, this book is a must have for all aspirants who want to start their career in data science or machine learning. What I feel missing is real life programming exercise at the end of each chapter. Also author could have taken one or two real life data science problems and could have solved the problem using the tools and techniques described in the book. This real life problem solutions could act as an example for aspirants of data science or machine learning jobs to learn and understand the needs of industry better. Even though, we would certainly say the book has done a fantastic job in making industry ready for new aspirants who wish to start their career in data science or machine learning.
We thank Manning Publications for providing with the opportunity to review this book.
GanitCharcha is delighted to receive from Manning Publications 35% discount code (good for all the Manning products in all formats) for our readers.
The discount code is : blgach21
One free eBook code for Math and Architectures of Deep Learning will be given to a lucky reader.
Link of the book :http://mng.bz/2jD8
One who is interested in taking part in the lucky reader contest will require to send us an email at email@example.com stating why he/she wants to read the book 'Math and Architectures of Deep Learning' by Krishnendu Chaudhury. Also share with us the full name of the person in the email. Subject line of the email should mention 'Math and Architectures of Deep Learning'. We will number each email sender as an integer starting from 101. So, the first email received will be numbered as 101, second email received will be numbered 102 and so on. We will accept email till 5th January, 2022, 6 PM IST. No email received after that will be considered for the lucky reader contest.
Min Number (which is 1) and the Max Number (the number assigned to the last email received within the stipulated time + 100) will be passed to the https://www.random.org/ to generate a random number within the range. Generating random number using this process is continued until we get a random number for which there is an email which is already labelled with that number. The sender of that email will receive from us the free eBook code for the book 'Math and Architectures of Deep Learning' . On 10th January, 2022 – we will publish the name of the lucky reader here. The free eBook code will be mailed to the lucky reader by 14thJanuary, 2022.
Winner of the luckey reader contest is Amit Kumar Nayak.