BUG: typeerror: datetime64 type does not support sum operations

[muazhari]

Pandas version checks

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

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

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

Reproducible Example

item_transactions_df = pd.DataFrame([value.dict() for value in item_transactions])

# item_transaction_df.dtypes 
# quantity_after            float64
# account_id                 object
# item_id                    object
# created_at         datetime64[ns]
# quantity_before           float64
# id                         object
# timestamp          datetime64[ns]
# updated_at         datetime64[ns]
# dtype: object

grouped_inventory_controls_df = item_transactions_df \
            .sort_values(by=["timestamp"], ascending=True) \
            .groupby(by=pd.Grouper(key="timestamp", freq='1D')) \
            .first()

resampled_item_transactions_df = grouped_inventory_controls_df \
            .resample(request.resample) \
            .sum()

Issue Description

Displayed error message:

typeerror: datetime64 type does not support sum operations

However, in version pandas==1.5.3, that error doesn't show up.

Expected Behavior

Could do group by and resample normally.

Installed Versions

INSTALLED VERSIONS

commit : 478d340
python : 3.9.13.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.22621
machine : AMD64
processor : Intel64 Family 6 Model 183 Stepping 1, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : English_United States.1252

pandas : 2.0.0
numpy : 1.21.5
pytz : 2022.1
dateutil : 2.8.2
setuptools : 63.4.1
pip : 22.2.2
Cython : 0.29.32
pytest : 7.3.0
hypothesis : None
sphinx : 5.0.2
blosc : None
feather : None
xlsxwriter : 3.0.3
lxml.etree : 4.9.1
html5lib : None
pymysql : None
psycopg2 : 2.9.6
jinja2 : 2.11.3
IPython : 7.31.1
pandas_datareader: None
bs4 : 4.11.1
bottleneck : 1.3.5
brotli :
fastparquet : None
fsspec : 2022.7.1
gcsfs : None
matplotlib : 3.5.2
numba : 0.55.1
numexpr : 2.8.3
odfpy : None
openpyxl : 3.0.10
pandas_gbq : None
pyarrow : 10.0.1
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : 1.8.1
snappy :
sqlalchemy : 1.4.41
tables : 3.6.1
tabulate : 0.8.10
xarray : 0.20.1
xlrd : 2.0.1
zstandard : None
tzdata : 2022.7
qtpy : 2.2.0
pyqt5 : None

[jbrockmendel]

In 1.5.3 you should have gotten a warning that silently dropping non-numeric columns in reductions was deprecated. you need to exclude the datetime64 column(s) before calling .sum

[muazhari]

In 1.5.3 you should have gotten a warning that silently dropping non-numeric columns in reductions was deprecated. you need to exclude the datetime64 column(s) before calling .sum

I need to process the timestamp column with the datetime64 type. How do I do that if I drop the column? Are there any other methods to do .sum() with datetime64 type columns?

[jbrockmendel]

when you do this in 1.5.3 you should see the column was getting dropped before. doing .sum() with datetime64 is gibberish

[muazhari]

In 1.5.3 you should have gotten a warning that silently dropping non-numeric columns in reductions was deprecated. you need to exclude the datetime64 column(s) before calling .sum

Do you mean that I need to drop datetime64 type columns first before processing it?

[jbrockmendel]

before calling .sum() at least, yes

[CabartPascal]

Having the same issue, looking into this since hours. How do I drop all datetime columns automatically? I have a dataset with 3000 columns and mixed datatypes of all kinds, is there no simple way to do it?

[jbrockmendel]

obj.select_dtypes(exclude=dtypes_to_exclude)

[phofl]

You can also set numeric_only=True. Closing since this behaves as expected

[Giwa-ibrahim]

I aslos ran into the same issue but using .sum(numeric_only=True) works perfectly.