Huge thanks to Eli Stevens, Luca Antiga, Thomas Viehmann for all their effort. In my view, this book is not only the perfect starting point for someone who wants to understand the fundamentals of PyTorch but equally good for advanced practitioners as the knowledge gained from this book can really help in writing neat and efficient PyTorch code .
A huge thank you to Eli, Luca and Thomas for writing such an amazing book!
I loved the detailed explanations of theoretical concepts, the accompanying code repository on GitHub and the amazing illustrations!
This book has truly helped me feel more confident writing PyTorch code. This is one of those books that I won’t just read and put to rest, but keep around on my desk always for ready reference.
This book is the way to a new world of DL . Since chapter 5, I have become a great fan of this book. These few weeks were a story of me taking my baby steps from ‘tensors’ to ‘Segmentations of Nodules’ which was not possible without Eli Stevens, Luca Antiga, and Thomas Viehmann.
They(The authors) had to transform themselves from being developers to basic ‘know nothing level’ and develop us as PyTorch practitioners, if they have not done it then I might have not seen PyTorch in the way that I do Today!! Thanks Again!
Thank you for putting together an absolutely amazing resource! The ramp up of concepts and the iterative process of modifying the code is such a great way to gently introduce PyTorch. I think this book is an amazing gateway into deep learning and I am certain will help a lot of new and existing DL practitioners.
It has been such a fun read too, with the humor and really helpful visuals. Thank you Eli Stevens, Luca Antiga, and Thomas Viehmann for taking the time to put this together!
I started my deep learning journey almost one year and a half ago with tensorflow. Then, six months ago after watching the interview of Jeremy Howard with Lex Fridman I decided to move to fastai and consequently to pytorch. Probably because fastai does its job too well, before the start of this reading group almost two months ago, I’ve never spent much time to explore vanilla pytorch and its functionality.
Let me tell you that I found your book the perfect companion (along with Sanyam) to assist me throughout this adventurous journey. The notions described in the first part, such as writing CNNs and training loops from scratch, have helped me to better approach repos on github and tailor them to my use cases. Let me also mention the fact that the author of the first part, Luca Antiga, is Italian like me makes me really happy and hopeful for the future of our country because unfortunately in Italy we’re lagging behind in many aspects in the technology development.
The second part was also an insightful reading because it taught me how to tackle a real business problem. I come from an unrelated educational background and I have to admit that I spend 99% of my time with colab notebooks. That’s why I think that the idea to structure the code for part 2 in .py files was a nice touch. Most of the repos on github are structured in a similar way and I always struggled to get my head around them. I’m sure that form now on it’s going to be a little bit easier.
Finally, the possibility to listen to and exchange ideas with Thomas Viehmann was also a great experience.
I hope the success of the book will convince Manning to fund a second volume because Eli, Luca and Thomas have done an excellent job with this first book and there is still A LOT to cover and I’m eager to learn .
I wish you all the best in life and for current and future projects.
P.S.
I also wanted to give a special thank to Sanyam and the WANB community to organize the reading group and accompany me throughout this journey. Thanks!
To me, this is the only time I found myself laughing and smiling at a Deep Learning book!
Since the beginning of my DL journey, I would often flock to the PyTorch docs, which I’ve studied enough to observe them grow from just having “60 min blitz” to now so many resources, yet it always felt like some dots were missing.
I sincerely think it was this book that was missing and really fills in the gap.
I still would have to revisit the book to sharpen my skills, but I sincerely thank Eli, Luca and Thomas for such an incredible resource and to the amazing W&B community for putting up with me during the sessions for 10 weeks! It was an honor leading the study group and making a very tiny contribution towards the amazing book!
Hey everyone, I’m personally beyond humbled by all the love . We had zero expectations around the reception of the book, so this is the best outcome ever possible.
I’m also blown away by the fact that you voluntarily decided to be subjected to our dad jokes for a few Sundays, eheh.
A big thank to @bhutanisanyam1 for being an early supporter of our work and to W&B for making this all possible. Best wishes to the community, looking forward to seeing great things being built! GO BUILD!!
and I’m eager to learn 
.
. We had zero expectations around the reception of the book, so this is the best outcome ever possible.