Worth learning OOP as a Data Scientist?
Hello blind. I’m less than a year into my new role as Data Scientist with no other technical background. I create, tune, analyze models in functions and nested functions. They get me the results I want and I just continue to improve it from there. My peers are satisfied in my coding (aside from adding too much info and cleaning up the code). I think there are lots of room for improvement in coding efficiency and thinking about the next skills to learn/improve on. I see OOP is something more advance and definitely not my strong suite. I only know Python and R, so no other programming experience as well. I know coding is just part of the role, but I just want to see how much effort I should put into learning this new skill and changing the way I code.Better to start now than later in my career. For current DS & MLE, do you use OOP much in your coding? Would you say one style is better than the other for this work? Or is the output more important? Or am I thinking about this incorrectly? Please explain. I’m still a noob. For SWE & other developers, how do you feel about this topic? What are your interactions with DS in the company in terms of their codes? Does it matter to you if a DS utilizes OOP or not? TC 80k
I would keep it low on priority.
Learn oop if you want to write production code for your models. If you don't use oop, your prod code will be significantly less maintainable and swe's will be horrified by what you hand to them. Oop also helps you understand and make better use of the libraries you use for model development, since they are all oop.
Makes sense. Much appreciated
It’s good to know for reproducible code but low on the list if you want to priortize other skills imo