Add quantile method to groupby object

[huard]

Dataset and DataArray objects have a quantile method, but not GroupBy. This would be useful for climatological analyses.

[tommylees112]

This is an awesome addition thankyou! I updated my xarray to V:0.12.2 but I am still getting an error with dataset objects

a = np.ones((400)) * 10
a[3:7] = 0.2
a[10:13] = 0.2
p = np.repeat(a, 25).reshape(400, 5, 5)
lat = np.arange(0, 5)
lon = np.arange(0, 5)
time = pd.date_range('2000-01-01', freq='M', periods=p.shape[0])
d = xr.Dataset(
    {'precip': (['time', 'lat', 'lon'], p)},
    coords={
        'lon': lon,
        'lat': lat,
        'time': time
    }
)

d.groupby('time.month').quantile(q=0.1)

gives the error message

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-38-276cfe319286> in <module>
----> 1 d.groupby('time.month').quantile(q=0.1)

AttributeError: 'DatasetGroupBy' object has no attribute 'quantile'

Whereas for the DataArray it works fine.

d.precip.groupby('time.month').quantile(q=0.1)

Out[39]:
<xarray.DataArray 'precip' (month: 12)>
array([10., 10., 10., 10., 10., 10., 10., 10., 10., 10., 10., 10.])
Coordinates:
  * month    (month) int64 1 2 3 4 5 6 7 8 9 10 11 12

Is this the expected behaviour?

INSTALLED VERSIONS
------------------
commit: None
python: 3.7.0 | packaged by conda-forge | (default, Nov 12 2018, 12:34:36)
[Clang 4.0.1 (tags/RELEASE_401/final)]
python-bits: 64
OS: Darwin
OS-release: 18.2.0
machine: x86_64
processor: i386
byteorder: little
LC_ALL: en_US.UTF-8
LANG: en_US.UTF-8
LOCALE: en_US.UTF-8
libhdf5: 1.10.4
libnetcdf: 4.6.2

xarray: 0.12.2
pandas: 0.24.2
numpy: 1.16.4
scipy: 1.3.0
netCDF4: 1.5.1.2
pydap: None
h5netcdf: None
h5py: 2.9.0
Nio: None
zarr: None
cftime: 1.0.3.4
nc_time_axis: None
PseudonetCDF: None
rasterio: 1.0.17
cfgrib: 0.9.7
iris: None
bottleneck: 1.2.1
dask: 1.2.2
distributed: 1.28.1
matplotlib: 3.1.0
cartopy: 0.17.0
seaborn: 0.9.0
numbagg: None
setuptools: 41.0.1
pip: 19.1
conda: None
pytest: 4.5.0
IPython: 7.1.1
sphinx: 2.0.1

[huard]

The PR only covered DataArrays, not Datasets, so yes, this is expected.

[dcherian]

Well that was an oversight. Any interest in opening a PR @tommylees112 or @huard

[dcherian]

OK looks like the method was added to the DataArrayGroupBy class not GroupBy. I should've caught this at review. Just moving the function definition to GroupBy seems to make things work