TST : add test for groupby aggregation dtype #43915
Conversation
OlivierCavadenti
commented
Oct 7, 2021
OlivierCavadenti
commented
- closes Aggregation does not preserve dtype when function passed in dictionary #32793
- tests added / passed
- Ensure all linting tests pass, see here for how to run them
- whatsnew entry
| @@ -1327,6 +1327,16 @@ def func(ser): | |||
| tm.assert_frame_equal(res, expected) | |||
|
|
|||
|
|
|||
| def test_groupby_aggregate_directory(): | |||
There was a problem hiding this comment.
can you parameterize over of the reduction kernels (use reduction_func)
and parameterize using frame_or_series
There was a problem hiding this comment.
Hi, thanks for the review.
I add reduction_func parameters and I try with frame_or_series parameter but it's seems that using a dictionary on a series for aggregation raise a "nested renamer is not supported" (seems deprecated in #15931 but now raise error).
Say me if I miss something (it's my first contribution).
jreback
added
Dtype Conversions
|
https://github.com/pandas-dev/pandas/pull/43915/checks?check_run_id=3885731836 fail, I check but I dont understand why in my code impact that. |
| @@ -1327,6 +1327,27 @@ def func(ser): | |||
| tm.assert_frame_equal(res, expected) | |||
|
|
|||
|
|
|||
| def test_groupby_aggregate_directory(reduction_func): | |||
| # GH#32793 | |||
| if reduction_func not in ["corrwith", "nth"]: | |||
There was a problem hiding this comment.
just do a return if the reduction_func is in rather than this big if
| if reduction_func not in ["corrwith", "nth"]: | ||
| obj = DataFrame([[0, 1], [0, np.nan]]) | ||
|
|
||
| obj.convert_dtypes() |
There was a problem hiding this comment.
what are you expecting this to do?
|
thanks @OlivierCavadenti |
