While writing my master thesis in electrical engineering (pure matlab based), I found the best way was to go about 2-3 days a week to the shared computer room from the department where other students where working. Most of those days I didn't get as much work done in terms of writing compared to when I was at home, but many days I got new ideas, algorithms or a new perspective by hearing other students talk about a different but related topic or discussing a problem with a college. So for me it was the best way to have both worlds, long phases of time at home for writing and implement stuff and quick interrupted time at the university for some quick code trying new stuff, discussing etc. It was commonly accepted that you are no as productive in the computer room and no one would blame you if you just picked up your notebook and went to a room next door for an hour or so to get stuff done.