✔️ Linting Passed

All linting checks passed. Your pull request is in excellent shape! ☀️

_{Generated for commit: fa4c590. Link to the linter CI: here}

Hi @MarcoGorelli, thanks for the PR :) Though I agree that we can change some examples to use polars, maybe this particular example doesn't really show an advantage with respect to using pandas. Instead we can address #28341 (comment), where @glemaitre mentions that time lagged feature engineering seems to be a more natural place to introduce polars.

Thanks @ArturoAmorQ for your review!

The advantage I was thinking of here is that Polars is strict about dtypes, unlike pandas doesn't let you do arithmetic on object dtype. The result of using pandas here was that it silently produced a "nonsense" plot:

This has since been addressed (#28345, #28571, #28352) but my point is that it was by trying to use Polars that the issue came up and was resolved - so here the advantage isn't in speed or memory usage, but strictness

Anyway, happy to close if you think the time lagged feature engineering example is a better fit

I was actually happy with this PR. I think it's worth the change, I just need to dig into the changed code a bit.

Just to be clear, I am not against this PR. If @adrinjalali is happy with it, then maybe we can highlight the benefit with a comment in the code or as narrative text.