Hello! I wanted to fix the nbdev linting error on neuralforecast, but then my PR fails with:

nbdev_export.............................................................Failed
- hook id: nbdev_export
- files were modified by this hook

But I only modified

setting.ini

and the

ci.yaml

files. Do you know why, and how I can fix it? Thanks for the help!

If you run nbdev_export locally with the new env, what changes do you see?

Honestly, quite a bit of .py files are changed when I just run

nbdev_export

What do the changes look like?

They're not changes to the code. Just a string at the top of the files like:

"""The `core.NeuralForecast` class allows you to efficiently fit multiple `NeuralForecast` models for large sets of time series. It operates with pandas DataFrame `df` that identifies individual series and datestamps with the `unique_id` and `ds` columns, and the `y` column denotes the target time series variable. To assist development, we declare useful datasets that we use throughout all `NeuralForecast`'s unit tests.<br><br>"""

Can you push those changes to the branch so we can review? It may have to do with the updated nbdev version which may be exporting differently.

Please note that there was a reason nbdev was pinned, versions after 2.3.25 remove modules with auto in its name, it wasn't just an arbitrary decision

That is what I had thought so I had initially fixed this by pinning

fastcore

instead, but you had changed that to removing the pinning for

nbdev

. So should we keep the same version of

nbdev

across the

Nixtlaverse

or can each repo have it's own version?

only in the repos where thats a problem. the nixtla repo doesn't have a module with auto so it was fine to upgrade

ok understood. So having different versions of nbdev across the Nixtlaverse repos is not a problem from a documentation point of view then?

no