Edoardo Abati

Hey @MarcoGorelli :) this sounds fun, I'd like to help. I'll go through `fairlearn`'s modules and update you here

Please see below the scripts where `pandas` is used (based on `grep -rl --include=\*.py "import pandas" fairlearn/*`) I'll slowly update this comment with my findings. If anyone else is interested,...

Your comment made me run all the tests to double check and I indeed found something missing. 💡 Note to self (and to anyone who will help): always run all...

Oh yeah you are right. 😅 I don't know why my brain thought it was less "safe". `.to_list` is not need then. Updated the comment

Couple of consideration regarding the `metrics` submodule. (I'm learning `polars` and `fairlearn` thanks to this issue so I may miss things, very happy to be corrected :) ) - Most...

Questions out of curiosity (it is not needed for this effort) I was looking at the column names (because of [this](https://github.com/fairlearn/fairlearn/blob/main/fairlearn/metrics/_group_feature.py#L59)) I noticed that a `narwhals.Series` name is always a...

Still have to see if I can find a workaround, but one method missing is `.transpose()`. It is used [here](https://github.com/fairlearn/fairlearn/blob/main/fairlearn/postprocessing/_tradeoff_curve_utilities.py#L167)

Hi @ugohuche thanks for helping out 🙂 I'll update my comment with your name. I checked your commit and FYI I'm taking a different approach (maybe not the ideal one)....

Hi @TamaraAtanasoska , I'd be happy to help a bit too! You're planning to tackle each file in a separate PR, right? What do you think could be the easiest...

I'll work on `utils/input_validation`. I think it is also a good one to see if there is anything missing on narwhals side