I adore Denis Rothman's Transformers and the way things are written and explained. A 5-year-old can understand his explanation and a scientist can improve his work. When I see his interviews, I learn, on the one hand, about the ChatGPT and Google Gemini transformers and, on the other hand, how to treat clients, be a proactive human, and get a new perspective on AI. AI becomes fun, easy, and life- and perspective-changing. I read the second edition, and I can not wait to apply what I read in the 3rd edition. I mean, it is written as if it would answer the pain points of becoming a GenAI pro and maximize business and living in any circumstance. I already attended a Packt conference with Denis Rothman as a speaker in October last year. With or without his opponents, GenAI is changing the world and will make an important difference. Thank you, Denis Rothman, for Transformers for Natural Language Processing and Computer Vision. I adore it.

This book is excellently presented. I find the material and examples stimulating and which has led to a million more questions. It a very complicated area to study but this book covers the topics very well. The code works well. Great book. Many thanks. I am still digesting this fabulous book.

If you're aspiring to become an expert in NLP or Generative AI, this book is an excellent resource. It provides a clear, step-by-step explanation of NLP models, making complex concepts easy to grasp through practical examples and Python code. . Starting with foundational models, the book introduces the architecture of Transformer, BERT, and RoBERTa, followed by an in-depth exploration of the GPT models which are the Generative AI revolution. The book also delves into image processing and computer vision. Additionally, the questions at the end of each chapter further enhance understanding and engagement with the material.

for those who can read, I can definitely say that this new third edition provides a fresh look at both the transformers themselves and the current environment in which they exist.A valuable resource to refresh our knowledge and inspire us to take the next stepsmy personal selection of what I appreciated in this third edition after about ten days of perusing, reading and note-takingthe emergence of new roles:* The role of AI professionals* The future of AI professionals* What resources should we use?* Guidelines for decision making* Chapter 3: Emergent vs. downstream tasks: The Unseen Depths of Transformers* Chapter 7: The Generative AI Revolution with ChatGPT* Chapter 12: Towards Syntax-Free Semantic Role Labelling with ChatGPT and GPT-4* Chapter 16: Beyond Text: Vision Transformers at the Dawn of Revolutionary AIRothman writes that this book is for data analysts, data scientists, and machine learning/AI engineers who want to understand how to process and interrogate the increasing amounts of speech and image data. Most of the programs in the book are Colaboratory notebooks. All you need is a free Google Gmail account and you can run the notebooks on the free Google Colaboratory VM.Context of my interest in this field: Shortly after the public release of ChatGPT in November 2022, Bill Gates described it and other LLMs as "as important as the PC, as important as the Internet". Jensen Huang, CEO of Nvidia, said ChatGPT was "truly one of the greatest things ever done for computing". Geoffrey Hinton, a Turing Laureate, said, "I think it's comparable in scale to the industrial revolution or electricity - or maybe the wheel. Perhaps that is why many of us need a qualified, updated context.I can definitely say that this new third edition gives a qualified context and fresh look at both the transformers themselves and the current environment in which they exist.and yet, the term "Computer Simulation" is far more accurate as an umbrella term than any characterization of machine software("AI," "LLM," "Generative AI," etc.).Rothman's profile shows that he has been designing and developing computer simulation software for decades in various forms: rule-based, expert systems, ML agents, DL agents, the first transformer models, and now trending Generative AI for NLP and Computer Vision. all these algorithms boil down to "computer simulation", no more, no less. They are toolss that are here for us to make "simulations" to enhance our abilities as a scientific calculator does.Who this book is for: Anyone who regularly works with LLMs professionally (e.g. data scientists, machine learning engineers, AI researchers, etc.) or anyone already familiar with natural language processing (NLP) who wants to take a deep dive into transformers.Another reviewer rightly wrote: Who this book is not for: Anyone with little to no knowledge of NLP, machine learning, or Python programming (i.e. the "casual" reader). This book is dense (in the sense of Clifford Geertz‘ thick description that helps us increase our understanding on both on a theoretical and a practical level). I still have a lot to think about.And I have to admit that I have not yet fully grasped all the emerging possibilities and food for thought that the book has triggered or will trigger as I re-read and explore the code provided.

The transformer architecture was introduced by Google in 2017, and almost instantly revolutionized the field of natural language processing (NLP), and to some degree also that of computer vision. This book is a comprehensive and practical guide to the transformer architecture, on which modern LLMs are based, and its applications in NLP and computer vision.The book does a wonderful job in providing detailed and clear descriptions of a wide range of important topics in NLP, such as the fundamentals of the transformer architecture, model pre-training, fine-tuning, tokenization, and embeddings. Notable applications of LLMs are also covered in detail, and include summarization, translation, etc. Modern generative AI methods are also very nicely covered, both in NLP and in computer vision (e.g., ChatGPT, Stable Diffusion, and the like). The accompanying GitHub repo is also very helpful, and greatly assists in reinforcing the concepts presented in the book.This comprehensive and unique guide will benefit any researcher, data scientist, machine learning engineer, or software engineer interested in building and understanding modern NLP and LLMs, as well as modern methods in computer vision. Prior familiarity with machine learning concepts, as well as with the Python programming language, would be helpful to get the most out of this book.Highly recommended!