BUG: df.agg(sum, axis=1) uses different method than when axis=0

[topper-123]

• closes BUG: df.agg(sum, axis=1) gives wrong result when Nan value is in frame #21134
• xref ENH: add np.nan funcs to _cython_table #21123
• tests added / passed
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

This is a splitoff from #21123, to only fix #21134. #19629 will be fixed in a separate PR afterwards.

Passing builtins to df.agg is ok when axis=0, but can give wrong result, when axis=1 when NaNs are supplied.

Explanation

Passing the functions in SelectionMixin._cython_table to df.agg should defer to use the relevant cython functions. This currently works as expected when axis=0, but not always when axis=1.

The reason for this difference is that df.aggregate currently defers to df._aggregate when axis=0, but defers to df.apply, when axis=1, and these give different result when passed funcions and the series/frame contains Nan values. I've solved this by transposing df in _aggragate when axis=1.

The tests have been heavily parametrized, helping ensure that the various ways to call df.agg now give correct result.

[pep8speaks]

Hello @topper-123! Thanks for updating the PR.

Cheers ! There are no PEP8 issues in this Pull Request. 🍻

Comment last updated on May 27, 2018 at 10:02 Hours UTC

[topper-123]

I’ve thought of a couple issues that should be tested. I mark this a WIP untill this is done.

Temporarily closed.