The need for foundational knowledge vs practical knowledge in ML [D]

MS statistics here. I did my undergraduate decrees in pure math and statistics. I find that due to my pure math background I’ve gotten in the habit of looking for the absolutely rigorous book for whatever it is I want to learn rather than jumping to something practical. I know how to program in Python and R, but just knowing how to fit a model doesn’t do it for me.

An example of this is when I started learning NLP …

book knowledge learn machinelearning math practical something statistics undergraduate