SeriesGroupby.nunique raises an IndexError on empty Series

[mfixman]

Code Sample, a copy-pastable example if possible

In [18]: b = pandas.Series()

In [19]: g = b.groupby(level = 0)

In [20]: g.nunique()
---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-20-fbbfc3108eac> in <module>()
----> 1 g.nunique()

/usr/local/lib/python2.7/dist-packages/pandas/core/groupby.pyc in nunique(self, dropna)
   2693 
   2694         out = np.add.reduceat(inc, idx).astype('int64', copy=False)
-> 2695         return Series(out if ids[0] != -1 else out[1:],
   2696                       index=self.grouper.result_index,
   2697                       name=self.name)

IndexError: index 0 is out of bounds for axis 0 with size 0

Expected Output

0

output of pd.show_versions()

INSTALLED VERSIONS
------------------
commit: None
python: 2.7.6.final.0
python-bits: 64
OS: Linux
OS-release: 3.13.0-79-generic
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: en_US.UTF-8

pandas: 0.17.1
nose: 1.3.1
pip: 8.0.2
setuptools: 20.1.1
Cython: None
numpy: 1.10.4
scipy: 0.13.3
statsmodels: 0.5.0
IPython: 4.1.1
sphinx: None
patsy: 0.2.1
dateutil: 2.4.2
pytz: 2015.7
blosc: None
bottleneck: None
tables: 3.1.1
numexpr: 2.2.2
matplotlib: 1.3.1
openpyxl: 1.7.0
xlrd: 0.9.2
xlwt: 0.7.5
xlsxwriter: None
lxml: None
bs4: 4.2.1
html5lib: 0.999
httplib2: 0.8
apiclient: None
sqlalchemy: None
pymysql: None
psycopg2: None
Jinja2: None

[jreback]

ok, pull-requests are welcome to fix

[thanasis2028]

Is this supposed to return 0 or an empty Series? I am fixing it right now but it seems awkward returning 0, while other calls of the nunique() return Series objects.

[jreback]

equiv to this, an empty Series should be returned. You should only do that null checking (the line where it errors), if the series has len

In [1]: s = Series([])

In [2]: s.groupby(s.index).sum()
Out[2]: Series([], dtype: float64)

[thanasis2028]

Thanks, made a pull request:
#12557
Edit: Output:

>>> reload(pandas)
<module 'pandas' from 'pandas/__init__.pyc'>
>>> pandas.Series().groupby(level = 0).nunique()
Series([], dtype: int64)
>>> pandas.Series([1],[1]).groupby(level = 0).nunique()
1    1
dtype: int64
>>>