vincent d warmerdam

I'm open to adding such a feature. I only wonder if we want to add parameters to the existing function or to create another function for the more elaborate use-case.

This sounds like an imputation combined with our [GroupedTransformer](https://scikit-lego.netlify.app/meta.html#Grouped-Transformation). I'm not 100% sure if the transformer has any notion of hierarchy. I do know that our grouped [predictor](https://scikit-lego.netlify.app/meta.html#Grouped-Prediction) does have...

I found it here: https://scikit-lego.netlify.app/api/preprocessing.html#sklego.preprocessing.RepeatingBasisFunction Does this not render for you?

@MBrouns do we still want to document the metrics we have listed here? https://github.com/koaning/scikit-lego/blob/main/sklego/metrics.py

It's currently linking to HTML yes. I was thinking of maybe running it as a local template but I didn't have the time to get around to that yet.

I've tried pointing to the HTML locally with Iframes but then there needs to be a server properly hosting them, which doesn't really play nice with jupyter security settings.

If you follow the url you'll see the js code hosted next to the html.

I'm open to making it an extension, but it'd be preferable if it would also run with users needing to install ipywidgets. I tried running it with python's `Template` class...

Adding extra information to an error message seems totally fine to me. Feel free to make the PR. @MBrouns, comments?

I once contemplated something like this, but then I discovered a small ocean of scikit-learn compatible packages that try to tackle the timeseries problem. Glancing at the [related packages guide...