Hi, I encountered a 'problem' when my program tried to groupby a dataframe with an empty (ie full of nan) column. Here a code sample with pandas 0.19.2 :

In [4]: a = pd.DataFrame({'a': [2,2,3,4,5], 
                          'b': [5,5,6,6,7], 
                          'c': [np.nan]*5, 
                          'sum': [1]*5})

In [5]: a
Out[5]: 
   a  b   c  sum
0  2  5 NaN    1
1  2  5 NaN    1
2  3  6 NaN    1
3  4  6 NaN    1
4  5  7 NaN    1

In [6]: a.groupby(['a','b','c']).agg({'sum':'sum'})
Out[6]: 
Empty DataFrame
Columns: [sum]
Index: []

In [7]: a.groupby(['a','b']).agg({'sum':'sum'})
Out[7]: 
     sum
a b     
2 5    2
3 6    1
4 6    1
5 7    1

I think we could throw a warning (or maybe an error) when this case is encountered in addition to return an empty dataframe.

INSTALLED VERSIONS ------------------ commit: None python: 3.6.1.final.0 python-bits: 64 OS: Linux OS-release: 4.4.0-75-generic machine: x86_64 processor: x86_64 byteorder: little LC_ALL: None LANG: en_US.UTF-8 LOCALE: en_US.UTF-8 pandas: 0.19.2 nose: 1.3.7 pip: 9.0.1 setuptools: 27.2.0 Cython: 0.25.2 numpy: 1.11.3 scipy: 0.18.1 statsmodels: 0.6.1 xarray: None IPython: 5.1.0 sphinx: 1.5.1 patsy: 0.4.1 dateutil: 2.6.0 pytz: 2016.10 blosc: None bottleneck: 1.2.0 tables: 3.3.0 numexpr: 2.6.1 matplotlib: 2.0.0 openpyxl: 2.4.1 xlrd: 1.0.0 xlwt: 1.2.0 xlsxwriter: 0.9.6 lxml: 3.7.2 bs4: 4.5.3 html5lib: None httplib2: None apiclient: None sqlalchemy: 1.1.5 pymysql: 0.7.9.None psycopg2: None jinja2: 2.9.4 boto: 2.45.0 pandas_datareader: None

Comment From: chris-b1

this is a duplicate of #3729. Some progress in #12607, but in the meantime would also probably take a PR for a better error/warning too.