Richard Shadrach

@mattB1989 - what do you think of this: ``` def test_group_on_empty_multiindex(transformation_func): # GH 47787 # Ensure empty vs non-empty frame/series gives same results df = DataFrame( data=[[1, Timestamp("today"), 3, 4]],...

The 32-bit tests are failing for diff and shift: ``` tm.assert_equal(result, expected) E AssertionError: Attributes of DataFrame.iloc[:, 0] (column name="col_3") are different E E Attribute "dtype" are different E [left]:...

Thanks for all the work here @mattB1989!

Agreed @mroeschke. I've been tagging deprecation issues where the conversation makes it clear there is no consensus as "Needs Discussion". There are currently only 46 issues that are tagged deprecation...

I believe this issue is meant only for methods that take a UDF (e.g. agg, apply, transform), is that correct? For other methods (e.g. sum, mean, fillna) I think there...

@jorisvandenbossche > But, for some of our groupby methods, we also internally basically use the UDF path and apply some method, and so those could still give errors / inconsistencies...

@jorisvandenbossche > Maybe we could also open a dedicated issue to discuss this specific issue for groupby with UDFs, and keep this as a general overview issue for working on...

This seems to me to be an example where `apply`'s inference cannot succeed in all cases. In such a case, I would typically suggest using `groupby.transform` instead as it no...

> my understanding was that `transform` expects a function that does not modify the shape or is broadcastable to the original shape of the group's data frame... This is currently...

I'm +1 on See Also and even an explicit note on when the Styler should be used (no surprise: adding styles to Excel output), but I don't think we should...