Approaching (Almost) Any Machine Learning Problem

The kind of book I was looking for. Not much mathematics not much beating around the bush. Straight code to show you how it's done. That's it. The real deal.I wanted to give it 4.5 stars if that was possible due few minor issues highlighted by some of reviewers. But those are minor points. No book is perfect for everyone. Probably the author can make some minor tweaks in the next edition. But for me, this perfect. So much to learn from it. Amazing value.

This is a very solid machine learning book with a strong focus on implementing things in Python/sklearn/PyTorch and how to do things in practice. I bought the book on the strength of the author's Kaggle success (well-known Kaggle Grandmaster) and excellent YouTube videos. I was not disappointed and got what I expected. I debated endlessly with myself whether this is four or five stars, because I would have liked a bit more depth/rationale for approaches (of course, that's a trade-off vs. the book length - and yes, I realized the cover text did tell me, but also things like pro-and-con of some methods might have been expanded), I had hoped even more types of problems/recent ideas would be discussed (don't get me wrong, a wide range of critical topics is covered really well - we are more talking about whether EfficientNet should be discussed in addition to ResNet, or whether more should be said on alternative loss functions like focal-loss, or the latest improvements over U-Net etc.), I wondered whether some topics were glossed over a little/downsides of some things not really discussed, and some minor formalistic complaints (no index, book title on the page headers instead of chapter titles, personally I might have liked the table of contents to show things below the chapter). Perhaps the length of some bits vs. other bits felt a tiny bit uneven (I guess this is just not an academic text-book that got polished over and over - which also makes it an accessible and a very pleasant fun read).But enough on the minor quibbles, here's some things that I really liked:1. Covering cross-validation really early on before starting with any supervised models: Definitely the right choice, excellent decision! In industry, too many people are sloppy with their model evaluation, emphasizing it this much is something people should be exposed to in more books. In fact, overall there's some pretty good didactic choices in how the material is arranged.2. I had not been sure about the approach of putting so much code directly into the book, but it works (it helps that the code is very readable/well-explained).3. Without trying to directly be a book primarily about PyTorch, it's actually a pretty good bare-bones starter pack for PyTorch (even if I prefer the more in-depth book by the fast.ai team).4. I like the author's writing style. Don't expect formalistic statistics or computer science text book language (and most definitely don't expect theorem - although some papers are referred to, but mostly when the author feels it would be worth your time to read them, there is no attempt to provide a reference for every statement or give a reference to who invented what method when). To get a feel, watch some of the authors YouTube videos - to me the book is a more concise, more polished product with a little bit more explanation of methods/why we use them (as said, maybe a little less than I prefer, but that's taste).4. Production value: The book is independently published, but with the exception of some of the complaints I mentioned above, I was very happy with design/layout/paper quality etc.

Nice code explanations to get into ML

I don’t normally buy books related to coding for the main reason that they usually become outdated quickly and end up taking up space on my bookshelf, never to be opened again.This book is different though. This is not about learning a specific package or for doing a specific task. This book aims to cover the fundamentals of machine learning regardless of what the task is. The (Python) code examples are easy to follow and use packages with mature APIs (e.g. scikit-learn, pandas, PyTorch) so the code examples will still work for years into the future. It also serves as a great reference for things which are often succinctly explained in a couple of paragraphs (e.g. the explanation of the AUC metric was probably one of the neatest I’ve seen). The other great thing about the writing style is that the theory and equations are intentionally kept to the bare minimum (you can find all of this elsewhere) so that you can get to building things as quickly as possible.If you are new to ML you should buy this book if you:> Have completed some MOOCs and are getting started with your own projects or Kaggle> Are wondering how that ML code that you copied on Stack Overflow/GitHub actually works> Prefer to learn quickly by doing rather than sitting through weeks of classesIf you have ML experience you should buy this book if you:> Started your ML journey through deep learning (e.g. fast.ai) and you want to firm up your ML fundamentals> Are coaching/mentoring less experienced colleagues and you need some succinct definitions/code examples of concepts like cross-validation or a certain metric> Want to improve your “coding style”, e.g. write clearer more maintainable codeI’d definitely recommend this book and it’s something I’ll probably find myself flicking back through for a while. I also really like the cover which has an awesome design and comes in a matt finish which feels nice in the hands while reading. Originally I thought it would pick up fingerprints very quickly but after owning it for several weeks it still looks great.

Really nice book that gives you the right perspective to approach a ML problem

Although not diving into theoretical details of machine learning (which is not the book’s intention), for practical aspects this is a very good book.