Ask HN: How do you organize your data science project files?
The more I do data science, the more I see the value of keeping as much information as I can about the modeling process during a project: all the metadata, data, code, models, predictions, discoveries from EDA, and decisions made along the way.
However, there are many ways storing all that information: many files in directories, Excel files to track the metadata on the models, pickled models, github repos, sqlite databases, etc.
I know that different organizations and teams have their own standards, but if you were to create a project from scratch, with your own rules, how would you organize it, and what tools would you use to do that?
No comments yet.