notebooks submissions troubleshooting faqs

Notebooks Submissions / Troubleshooting and FAQs

My notebook needs to access a dataset. What are my options?

For the Reproducible Documents submission category, the idea is to keep the notebook as reproducible and self-contained as possible, which typically means ensuring that all steps of the workflow are implemented within the notebook itself. The idea is that the data should either be small enough to fit in your repository on GitHub, or small enough that you can load the dataset from a remote source

Loading the data from a remote source can be done in several ways:

• If you choose to use a Python or R function to fetch the dataset, just remember to include that library in the list of dependencies you specified in your environment file (see the Repository checklist page for reference)
• You may also use a command line tool such as wget. If you are planning to submit a reproducible document (which compiles to static HTML), you can refer to the list of preinstalled software available on the runners that will be used to validate the submission

If you are planning to submit an interactive notebook (which will be launched interactively with Binder) you can refer to this very helpful page on GitHub called “How to get data into your Binder”