No you're right, I read/skimmed it cover to cover, but that doesn't do that book justice. I didn't do a single exercise and I really focused on the sections tha

No you're right, I read/skimmed it cover to cover, but that doesn't do that book justice. I didn't do a single exercise and I really focused on the sections that were relevant to what I was doing (Random Forests, testing/validation). The author recommends maybe 4 or 5 chapters as required reading in the forward so I read those much more thoroughly, but again, skipped all of the exercises and the appendices. At the end of the day though, I'm not going to be implementing these algorithms, I'm just using existing libraries so I couldn't really justify the year+ of effort to go in depth on everything. That book is also a summarization of a lot of techniques. If you wanted more depth, I wouldn't think Elements of Statistical Learning is the book for it. I'm hoping to get fill in some of those gaps through reading Optimization Models a bit more closely.