Modern Computer Vision with PyTorch: Explore deep learning concepts and implement over 50 real-world image applications

The authors have done a fantastic job in writing this book.I bought this book specifically to implement object detection and face recognition systems. The online notebooks are succinct and very clear. The library, torch_snippets, created by the authors is very useful.Very happy with my purchase. Wish the authors great success in their careers and future writing!

Great book! You get a lot of tutorials on the internet but nothing beats a structured learning path .

Very practical. Just finished the first 2 chapters and enjoying it so far.

I have been invited to write a review for this book, I have generally reviewed, I found this is a very good book if you are a beginner or interested in the deep learning field with using python. Most of this book introduced experiments or code snippets based on Pytorch which is a very popular deep learning library in academic research and industries. This book can equip you with an overview of the necessary knowledge along with practical implementations, after reading this book, it expects that at least you could solve basic problems or even harder ones in your tasks and you would feel be comfortable acquiring new knowledge in a fast way.Enjoy reading!

I was invited by the publisher to review this book and provide my thoughts on that. Overall, I found the book to be very informative and well laid out. It takes you through the very basics of backpropagation to advanced implementations in PyTorch. The book assumes that the reader does have an understanding of Python at the very least, which I think beginners in deep learning should be aware of. Please find my honest thoughts on the book as follows:What I liked:I felt the book is very well structured and compiled. Unless you're looking for something very very specific, you'd be able to find techniques/implementations for any and all types of problems you are working on. They cover algorithms and implementations of basic neural networks, all the way upto RNNs and reinforcement learning with PyTorch. The breadth covered by this book on the number of techniques and algorithms is really amazing. Not only that, each of these topics are covered with amazing depth as well, making the book a must have for someone just starting out with PyTorch. The progression in difficulty of the book, also makes it beginner friendly.What I would like to be there:I would love to see 2 things added to this book:1. Tensorboard setup with PyTorch, and usage. Although it's trivial, it would take some getting used to in setting up tensorboard with PyTorch. It's fairly common to use a graphical visualization tool like this while training and tracking your models. A small chapter, or a section in the appendix on how to setup tensorboard with PyTorch, and some examples to integrate that into your pipeline, would really be amazingly beneficial for any beginner/ intermediate user!2. PyTorch models can't generally be used directly in the industry because most embedded chips don't have compilers for .pth models. Most chip manufacturers (unless you're using nvidia GPUs itself), have compilers implemented for either tensorflow, caffe models or the cross compatible ONNX format. I think a small section on converting your models into ONNX would really be beneficial for anyone just starting out and targeting embedded/high performance applications on the edgeAll in all, I loved and enjoyed reading the book. It is a definite recommendation from my side for anyone who is starting out with PyTorch, or someone looking to implement advanced algorithms (more than just the basic feedforward networks or CNNs). Amazing work by Ayyadevara and Reddy! :)

This book is very well written and comprehensive! I agreed with another reviewer that the information about computer vision is quite scattered and this book contain all the required information in one place. This really saves me time to getting started with computer vision. Also the PyTorch examples contained in the book is very clean and handy. It also covers all major topics in computer vision. This definitely is a book for both beginners and advanced practitioners.

Ayyadevara and Reddy’s 'Modern Computer Vision with PyTorch' is a well constructed beginner to intermediate level text on working more efficiently and creatively with PyTorch in image analysis and CV techniques. The chapters are well laid out in a logical progression from neural network and PyTorch fundamentals to more advanced topics in neural network design and model implementation. In addition to the theory behind the concepts, the authors include numerous and easy to follow examples, and make all code available via Github. This practical, hands-on approach allows readers to more easily grasp complex ideas and techniques. Supported by the innovative industry experience of both authors, the inclusion of a chapter on moving a model to production, a topic often overlooked, will be highly useful to the budding data scientist. This text acts as an excellent resource for those interested in both skill-building and -acquisition.

I found this book is very well written. The book is a self-contained starting with the basics, the fundamentals about artificial neural networks, fundamentals about PyTorch and several basic deep learning examples using PyTorch.The book then discusses important topics in computer vision, such as object classification and detection, image manipulations etc. The book also covers advanced topics, such as GANs, reinforcement learning, self-attention, few-shot learning etc. It is very handy that all these discussions come with concrete PyTorch examples.It is a very good book for both beginners and the practitioners who want to learn the most recent updates and advances.