DOC: axis=None does not aggregate along both axes for sum method

[Ishticode]

Pandas version checks

• I have checked that the issue still exists on the latest versions of the docs on main here

Location of the documentation

https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.sum.html

Documentation problem

The comment under axis parameter states:

For DataFrames, specifying axis=None will apply the aggregation across both axes.

However the following code seems to suggest otherwise.

import pandas as pd, numpy as np
a = np.arange(6).reshape((3, 2))
df = pd.DataFrame(a, index=list("abc"), columns=list("xy"))
print(df.sum(axis=None)) # or df.sum()

which returns

x    6
y    9
dtype: int64

and seems to sum just along the row as if we had specified axis="index".

In additon to this, the return statement for the sum docs linked above says scalar or series. If the above case does not return a scalar then I can't see when it will be scalar otherwise.

Note:

1. This behaviour is diffierent to numpy.ndarray.sum which would return a scalar value of 15.
2. This behaviour is also found amongst other similar df methods like mean.

Suggested fix for documentation

remove
For DataFrames, specifying axis=None will apply the aggregation across both axes.
or
make the methods match numpy behaviour which is perhaps a breaking change.

[jbrockmendel]

The current behavior is deprecated; in 3.0 explicitly passing axis=None will sum across both axes. A PR making the documentation more clear in the interim would be welcome.

[nox1134]

Hii @rhshadrach ! I'm new to open source and would like to work on this issue. How do I get started?

[selehadin-cyber]

Hi @rhshadrach

can you assign me this? can do a quick PR
thanks in advance

[Brooklynn29]

take

[nox1134]

@rhshadrach I've set up the environment and installed all dependencies, but I'm unsure how to edit docstrings. Should i create a new function and override the the existing sum method?

[salujaditi14]

@Ishticode , @jbrockmendel , @Brooklynn29 , Do you guys would like to look into this PR?

#54681

[haleematallat]

take

[natmokval]

I would like to work on it.