To be honest, I don’t think any of these books are really very useful. I’ve been doing data science professionally for over two years, and if I had to go back in time and advice myself to read some of the books on data science, it wouldn’t be any of these. I would recommend ESL or ISL (books and youtube lectures), Max Kuhn’s book, doing some live tutorials of scikit-learn, data.table and pandas “5-minute” vignettes, and just going straight into Kaggle.
Data Science at the Command Line comes close, though. I’ve looked at that book early enough, and it has some nice CL tools that were very helpful.
