How this book is different from the "The Hundred-Page Machine Learning Book" or "Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems"?
"The Hundred-Page Machine Learning Book" will teach you what ML is. It can be a treated as a light weight replacement for say an Andrew NG course. You wont learn anything practical / code something in PyTorch or Tensorflow but you ll understand what happens under the hood in these framework
"Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems" is an amazing introduction to implementing all ML ideas. I think there is a new PyTorch version. The accompanying notebook will get you started to a point where you can do some hobby projects.
This book surprisingly seems to fill an interesting gap explaining about how these ML systems are used in real life in large scale. I work for 1 of the FAANG companies & I can say that every chapter here would correspond to bread & butter of a team responsible for maintaining a large ML system say Recommendation / Fraud detection. The target audience would be someone who is interested to learn how to put a large end-end ML system together.
I would be very excited if there are practical examples on how to use this with MLFlow / KubeFlow / Sagemaker. Really excited to read this
I think that it's unfair of you just to assert this without giving some specific criticism of this book.
I am pretty interested actually as I am currently trying to write a similar book of my own! I can see a lot of difference between this and what I am trying to do, and may differences to "Hands on" but what in particular are you disappointed with here?
The book in the original post and the other book from the same author are short surface-level summaries of the field that don't go in depth at all and don't provide any meaningful new content. It's like someone took their personal notes, formatted them in a pdf and called them a book. It's an embarrasingly transparent cash grab or just something that helps the author's social media presence because they can now say that they "have written an ML book".
> It's like someone took their personal notes, formatted them in a pdf and called them a book.
Why is that such a bad thing? I like condensed, reader's digests versions of things. Not every text has to break new ground; some of it can be, well, purely educational.
Or do you dispute the educational value of the text?
Legitimately curious; I haven't read any of the books, but I actually like the way you described the author's text process.
I agree - "new material" is actually a concern in the context of a text book. I feel that synthesizing and presenting a coherent view of the communities best practice is a very valuable thing.
I think (but am very open to correction) that this is now a very hard thing for academics to do - the incentive for writing a text book is very low because they are not esteemed or counted as research? Writing a book like this is very hard.
IIRC, authors are usually not supposed to release the final version for a while due to some legal complications even if their publisher agrees with the idea itself. One of the workaround is releasing a draft version of the book, nearly identical to the final version. If this is the case, the book can be practically considered public.