Greetings!
I’m working as a researcher in a field very close to ML for about a year. I’m trying to make my projects as clean as possible in terms of code-writing, implementing W&B for experiment tracking, organizing everything in a particular structure, etc.
But recently I thought that despite my eagerness, my ML (Pytorch) pipelines are still custom-made, and I haven’t found any particular “The most classical” style guide (including experiment tracking, files, and classes organization) to stick with.
So my question is: are there any resources to go through if I want my code to look more like “end-to-end enterprise solution”?
3 Likes