PERF: `mean` method works too long with object type

[Tialo]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this issue exists on the latest version of pandas.

• I have confirmed this issue exists on the main branch of pandas.

Reproducible Example

import pandas as pd

df = pd.DataFrame(["hello world"] * int(5e6))
df.mean()

It concatenates million strings, which is basically to slow. But then it fails, because result string is not numerical. Maybe it would be right to check if it is numerical first?

Installed Versions

INSTALLED VERSIONS

commit : 0f43794
python : 3.8.17.final.0
python-bits : 64
OS : Darwin
OS-release : 22.5.0
Version : Darwin Kernel Version 22.5.0: Thu Jun 8 22:22:22 PDT 2023; root:xnu-8796.121.3~7/RELEASE_X86_64
machine : x86_64
processor : i386
byteorder : little
LC_ALL : None
LANG : None
LOCALE : None.UTF-8

pandas : 2.0.3
numpy : 1.24.4
pytz : 2023.3
dateutil : 2.8.2
setuptools : 56.0.0
pip : 23.0.1
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : 2.9.7
jinja2 : 3.0.3
IPython : 8.12.2
pandas_datareader: None
bs4 : 4.12.2
bottleneck : None
brotli : None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : None
numba : None
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : 12.0.1
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : 1.10.1
snappy : None
sqlalchemy : 2.0.20
tables : None
tabulate : 0.9.0
xarray : None
xlrd : None
zstandard : None
tzdata : 2023.3
qtpy : None
pyqt5 : None

Prior Performance

No response

[lithomas1]

Your DF is object dtype because of the strings.
I think a check might be moderately expensive in the case you have object dtype of e.g. numbers
(since you would need to iterate over every element to check if its numeric)

Also, note that the default value of numeric_only is False for mean. You might want to pass in numeric_only=True if you want to exclude the strings.

[Tialo]

I am not sure that numeric_only will work that way. Just checked

import pandas as pd

df = pd.DataFrame(["asd", 1, 4])
mean = df.mean(numeric_only=True)

this code gives mean = Series([], dtype: float64)
It skips not values, but columns.

So if it does not check each value, maybe it will be not expensive to just check dtype of a column?

[lithomas1]

Your DF is of object dtype though.

Anything can be in an object dtype ndarray which is why you can't just check onlythe dtype here.

[Tialo]

Oh, I see. Well if there is a way to check if columns has string, that would be great to know before, trying to concatenate millions of them.