DataFrame.gropuby().mean() incorrect result

[tinchoroman]

Anybody knows why I'm having different results when I apply the same operator to the same DataFrame but using groupby?
When using groupby , It returned negative values while all values are positive.

from pandas import DataFrame
df = DataFrame({"user":["A", "A", "A", "A", "A"],
                            "connections":[18446744073699999744, 4970, 4749, 4719, 4704]})

df.mean()

connections 3.689349e+18
dtype: float64

df.groupby("user")["connections"].mean()

user
A -1906546.0
Name: connections, dtype: float64

INSTALLED VERSIONS

commit: None
python: 3.6.5.final.0
python-bits: 64
OS: Windows
OS-release: 10
machine: AMD64
processor: Intel64 Family 6 Model 142 Stepping 9, GenuineIntel
byteorder: little
LC_ALL: None
LANG: None
LOCALE: None.None

pandas: 0.23.0
pytest: 3.5.1
pip: 10.0.1
setuptools: 39.1.0
Cython: 0.28.2
numpy: 1.14.3
scipy: 1.1.0
pyarrow: None
xarray: None
IPython: 6.4.0
sphinx: 1.7.4
patsy: 0.5.0
dateutil: 2.7.3
pytz: 2018.4
blosc: None
bottleneck: 1.2.1
tables: 3.4.3
numexpr: 2.6.5
feather: None
matplotlib: 2.2.2
openpyxl: 2.5.3
xlrd: 1.1.0
xlwt: 1.3.0
xlsxwriter: 1.0.4
lxml: 4.2.1
bs4: 4.6.0
html5lib: 1.0.1
sqlalchemy: 1.2.7
pymysql: None
psycopg2: None
jinja2: 2.10
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None

[WillAyd]

Can you try on master? Looks like an int overflow somewhere if still present investigation and PRs are always welcome

[tinchoroman]

Hi WillAyd, thanks for your prompt response. This is the first time I post an issue. Could you please explain little further what you exactly mean by "try on master" ? Thanks in advance!

[tinchoroman]

I've upgraded to latest version and the problem still persists. In the investigation line that WillAyd suggests, the same example whit float numbers worked fine.

df = DataFrame({"user":["A", "A", "A", "A", "A"],
           "connections":[18446744073699999744.0, 4970.0, 4749.0, 4719.0, 4704.0]})

df.mean()
connections    3.689349e+18
dtype: float64

df.groupby("user")["connections"].mean()
user
A    3.689349e+18
Name: connections, dtype: float64

df.mean()[0] == df.groupby("user")["connections"].mean()[0]
True