BUG/ENH: groupby.rolling.agg / Column(s) low already selected

[Earthson]

GroupBy.rolling.agg failed with Column(s) low already selected

data = pd.DataFrame({"stock": [1, 1, 1, 2, 2, 2, 2], "low": [10, 20, 30, 10, 30, 40, 80]})
data.set_index("stock", inplace=True)
# failed
data.groupby(level="stock").rolling(2).agg({"low": {"mean": "mean", "max": "max"}})
# works
data.rolling(2).agg({"low": {"mean": "mean", "max": "max"}})

This also hits groupby().resample.agg

In [13]: df = pd.DataFrame({"A": pd.to_datetime(['2015', '2017']), "B": [1, 1]})

In [14]: df
Out[14]:
           A  B
0 2015-01-01  1
1 2017-01-01  1

In [15]: df.set_index("A").groupby([0, 0]).resample("AS")
Out[15]: DatetimeIndexResamplerGroupby [freq=<YearBegin: month=1>, axis=0, closed=left, label=left, convention=e, base=0]

In [16]: df.set_index("A").groupby([0, 0]).resample("AS").agg(['sum', 'count'])
---------------------------------------------------------------------------
Exception                                 Traceback (most recent call last)
<ipython-input-16-5f1c18a8d4ac> in <module>()
----> 1 df.set_index("A").groupby([0, 0]).resample("AS").agg(['sum', 'count'])

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/resample.py in aggregate(self, arg, *args, **kwargs)
    339
    340         self._set_binner()
--> 341         result, how = self._aggregate(arg, *args, **kwargs)
    342         if result is None:
    343             result = self._groupby_and_aggregate(arg,

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/base.py in _aggregate(self, arg, *args, **kwargs)
    538             return self._aggregate_multiple_funcs(arg,
    539                                                   _level=_level,
--> 540                                                   _axis=_axis), None
    541         else:
    542             result = None

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/base.py in _aggregate_multiple_funcs(self, arg, _level, _axis)
    583                 try:
    584                     colg = self._gotitem(col, ndim=1, subset=obj[col])
--> 585                     results.append(colg.aggregate(arg))
    586                     keys.append(col)
    587                 except (TypeError, DataError):

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/resample.py in aggregate(self, arg, *args, **kwargs)
    339
    340         self._set_binner()
--> 341         result, how = self._aggregate(arg, *args, **kwargs)
    342         if result is None:
    343             result = self._groupby_and_aggregate(arg,

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/base.py in _aggregate(self, arg, *args, **kwargs)
    538             return self._aggregate_multiple_funcs(arg,
    539                                                   _level=_level,
--> 540                                                   _axis=_axis), None
    541         else:
    542             result = None

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/base.py in _aggregate_multiple_funcs(self, arg, _level, _axis)
    582             for col in obj:
    583                 try:
--> 584                     colg = self._gotitem(col, ndim=1, subset=obj[col])
    585                     results.append(colg.aggregate(arg))
    586                     keys.append(col)

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/base.py in _gotitem(self, key, ndim, subset)
    675                        for attr in self._attributes])
    676         self = self.__class__(subset,
--> 677                               groupby=self._groupby[key],
    678                               parent=self,
    679                               **kwargs)

~/Envs/pandas-dev/lib/python3.6/site-packages/pandas/pandas/core/base.py in __getitem__(self, key)
    241         if self._selection is not None:
    242             raise Exception('Column(s) {selection} already selected'
--> 243                             .format(selection=self._selection))
    244
    245         if isinstance(key, (list, tuple, ABCSeries, ABCIndexClass,

Exception: Column(s) B already selected

Problem description

It's not work as expected

Output of pd.show_versions()

INSTALLED VERSIONS ------------------ commit: None python: 3.5.2.final.0 python-bits: 64 OS: Linux OS-release: 4.4.0-53-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+0.g825876c.dirty
nose: 1.3.7
pip: 9.0.1
setuptools: 27.2.0
Cython: 0.24.1
numpy: 1.11.1
scipy: 0.18.1
statsmodels: 0.6.1
xarray: None
IPython: 5.1.0
sphinx: 1.4.6
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: 1.5.3
openpyxl: 2.3.2
xlrd: 1.0.0
xlwt: 1.1.2
xlsxwriter: 0.9.3
lxml: 3.6.4
bs4: 4.5.1
html5lib: None
httplib2: None
apiclient: None
sqlalchemy: 1.0.13
pymysql: 0.7.9.None
psycopg2: None
jinja2: 2.8
boto: 2.42.0
pandas_datareader: None

[jreback]

sophisticated .agg was never explicitly implemented on .groupby.rolling, so not surprising this doesn't work. if you'd like to debug please do.

In [12]: data.groupby(level="stock").rolling(2).mean()
Out[12]:
              low
stock stock
1     1       NaN
      1      15.0
      1      25.0
2     2       NaN
      2      20.0
      2      35.0
      2      60.0

In [13]: data.groupby(level="stock").rolling(2).max()
Out[13]:
              low
stock stock
1     1       NaN
      1      20.0
      1      30.0
2     2       NaN
      2      30.0
      2      40.0
      2      80.0

[mahnunchik]

Hi All!

I've faced with the same error: https://stackoverflow.com/questions/46350611/aggregate-group-by-rolling-window-keyerror-column-not-found-foo