Saving and persisting data in Kaggle

There are a lot of reasons you might want to save data from a Kaggle notebook. Maybe you’ve sanitized a dataset in Python, and now you need to pass it to a colleague who prefers to work in Excel. Maybe you’ve trained a useful machine learning model and want to use it across multiple projects. Maybe you’re four hours into a project, and your yoga class starts in half an hour, and you’re worried about losing all your work.

Whatever the situation may be, you need to make sure that your data is safe when your notebook session shuts down so you can pick up where you left off later. Unfortunately, there’s a lot of confusion and misinformation in the Kaggle discussion board about how to save data from a notebook and persist it across multiple notebook sessions. In this notebook, I hope to clear up some of that confusion. I’ll explain where and when Kaggle saves your data, how to persist data across notebook sessions, and how to load a notebook’s output into a new notebook or dataset. I’ll also provide a few tips and tricks to help you get the most out of your saved data.