Date Type Corrupting Other Types in Group-by/Apply

[gwpdt]

Code Sample, a copy-pastable example if possible

In [1]: import pandas as pd

In [2]: df = pd.DataFrame({'Number' : [1, 2], 'Date' : ["2017-03-02"] * 2, 'Str' : ["foo", "inf"]})

In [3]: df
Out[3]:
         Date  Number  Str
0  2017-03-02       1  foo
1  2017-03-02       2  inf

In [4]: df.groupby(['Number']).apply(lambda x: x.iloc[0])
Out[4]:
              Date  Number  Str
Number
1       2017-03-02       1  foo
2       2017-03-02       2  inf

In [5]: df.Date = pd.to_datetime(df.Date)

In [6]: df
Out[6]:
        Date  Number  Str
0 2017-03-02       1  foo
1 2017-03-02       2  inf

In [7]: df.groupby(['Number']).apply(lambda x: x.iloc[0])
Out[7]:
             Date  Number  Str
Number
1      2017-03-02       1  NaN
2      2017-03-02       2  inf

Problem description

When I change the type of the Date column to a Pandas datetime, it causes other columns' types to change in unexpected ways when doing a group-by/apply. Notice the contents of the "Str" column changes to a numeric type in the final group-by/apply (a contributing factor is probably that one of the elements is the string "inf"). The "inf" value has become inf, and the "foo" value has become NaN.

Expected Output

I expect the Str column to remain a string type, and contain the original strings. I.e.:

        Date  Number  Str
0 2017-03-02       1  foo
1 2017-03-02       2  inf

Output of pd.show_versions()

INSTALLED VERSIONS ------------------ commit: None python: 2.7.11.final.0 python-bits: 64 OS: Linux OS-release: 3.10.0-327.10.1.el7.x86_64 machine: x86_64 processor: x86_64 byteorder: little LC_ALL: None LANG: en_US.UTF-8

pandas: 0.18.1
nose: 1.3.7
pip: None
setuptools: 0.6
Cython: 0.24.1
numpy: 1.11.1
scipy: 0.18.0
statsmodels: 0.6.1
xarray: 0.7.0
IPython: 5.0.0
sphinx: 1.3.5
patsy: 0.4.1
dateutil: 2.5.3
pytz: 2016.6.1
blosc: None
bottleneck: 1.1.0
tables: 3.2.3.1
numexpr: 2.6.1
matplotlib: 1.5.1
openpyxl: 2.3.2
xlrd: 1.0.0
xlwt: 1.1.2
xlsxwriter: 0.9.2
lxml: 3.6.1
bs4: 4.4.1
html5lib: 0.999
httplib2: None
apiclient: None
sqlalchemy: 1.0.13
pymysql: 0.6.7.None
psycopg2: 2.5.4 (dt dec pq3 ext)
jinja2: 2.8
boto: 2.40.0
pandas_datareader: None

[jreback]

this is a duplicate of this: #14423

soln is pretty easy if you'd like to do a PR