PROGRAMMING PYTORCH FOR DEEP LEARNING

by Ian Pointer

Deep learning is changing everything. This machine-learning method has already surpassed traditional computer vision techniques, and the same is happening with NLP. If you're looking to bring deep learning into your domain, this practical book will bring you up to speed on key concepts using Facebook's PyTorch framework.Once author Ian Pointer helps you set up PyTorch on a cloud-based environment, you'll learn how use the framework to create neural architectures for performing operations on images, sound, text, and other types of data. By the end of the book, you'll be able to create neural networks and train them on multiple types of data.Learn how to deploy deep learning models to productionExplore PyTorch use cases in companies other than FacebookLearn how to apply transfer learning to imagesApply cutting-edge NLP techniques using a model trained on Wikipedia

Genres: Programming, Computer Science, Artificial Intelligence, Technology

Paperback

Reviews

[Niyousha · Rating 2 out of 5]

I wish the codes had more explanation, most of the time I felt lost. But it was a good book to start.

[Bulat Ibragimov · Rating 1 out of 5]

It is a superficial book that does not cover any topic in sufficient detail.
Chapters 2 and 4 are quite good because they are about pytorch.
Chapters 3, 5, and 6, instead of talking about working with pytorch, bump into superficial stories about neural network models.
I would be glad if chapters 7 and 8 were much more detailed.
Chapter 9 is completely useless, it just mentions different methods and technologies from the world of neural networks, but does not explain anything.

[Vladimir · Rating 4 out of 5]

Nach dem Buch "Praxiseinstieg Machine Learning mit Scikit-Learn, Keras und TensorFlow: Konzepte, Tools und Techniken für intelligente Systeme. Aktuell zu TensorFlow 2" vom Schriftsteller Aurélien Géron wollte ich eine andere Bibliothek für Machine Learning lernen. Pytorch kommt man nicht vorbei, wenn es um das Thema geht - deswegen habe ich mich entscheiden dieses Buch zu lesen.

Im Nachhinein kann ich sagen, dass die Reihenfolge von Büchern meine Meinung nach richtig gewesen war. Beim ersten Buch konnte ich das theoretische Teil von Machine Learning sehen, während hier wird es mehr auf die Beispiele aus der Praxis eingegangen. Für jemand, wie ich, der hobbymäßig mit Python programmiert, wird es wahrscheinlicher einfacher die Logik von Tensorflow zu verstehen, da Pytorch etwas abstrakter ist. Nichtdestotrotz, wie man im Buch sehen kann, ist es möglich die Aufgaben mit beiden Bibliotheken genauso gut zu erledigen.

Gefehlt haben mir die vertieften Unterkapitel bei den Bildklassifizierungen und CNNs. Etwas unerwartet und sehr praktisch fand ich das Kapitel über Debuggen der Modelle, worüber man leider viel zu selten spricht. Insgesamt schafft man eine gute Übersicht davon, wie man ein Machine-Learning-Modell programmieren, debuggen und in die Produktion einsetzen kann.

[Octavio Pavon · Rating 3 out of 5]

It's a book for someone that already has some skills in deep learning. isn't a book for beginners.
Anoter funny fact is that this book contains some code errors, but are not too dangerous, and if you're agile with coding, you'll notice fast. Despite of that, it helped me a lot moving from tensorflow to PyTorch. I'll recommend it if you want to do the same, isn't a book for learning deep learning, for that task i'd recommend "Learning Deep Learning - Magnus Ekman", "Machine Learning - Aurelien Geron", or Francois Choillet books. Regardless of that, this book makes you to build an AlexNet, it covers some history and gives you good tips for a general Deep Learning Development. Im applying an using this concepts right now with some of personal projects. Another fact, this book was released in 2019, so pytorch wasn't well developed yet, remember that pytorch released date was 2017, so, nowadays pytorch doesn't contains any of the limitations that the book have mentioned, take that in mind (aka, you could do with the same coding more)

[Gary Bake · Rating 2 out of 5]

Coming from Keras, I wanted a small book to learn pytorch. This sort of fits the bill, there are nuggets of pretty good insight in there. Unfortunately there are a ton of bad mistakes in there also. None of the code has been checked and there are references to paragraphs that have been taken out. It feels rushed with no quality checking done before publication.

[Abdul-Haleem Osama · Rating 3 out of 5]

The book offers a fast-paced overview of deep learning techniques but is marred by numerous typos and code errors. Some sections are overly abstract, making the material harder to follow. Overall, it needs improvement to be a reliable learning resource. 2.5 stars.

[Silentroar · Rating 3 out of 5]

One of the books for complete newbies. There is little value left after you've used Pytorch for some time.

[Rodrigo Pedrosa · Rating 5 out of 5]

It delivers what it sells.
It's a great PyTorch introductory book.

[Monirul Haque · Rating 2 out of 5]

Very poor quality book. Also, most of the codes don't work anymore.

[Ramtin · Rating 3 out of 5]

I previously used TensorFlow and keras and I have decided to learn PyTorch too. It is a good book for beginners with some backgrounds regarding neural networks. Only problem is that the book repo need to be updated because some of the scripts are not working and out of dated.

[Vinay Khobragade · Rating 4 out of 5]

The book is decent with regards to Pytorch for creating deep learning models. However, sufficient content is not available for deploying the applications, as the title suggests.