Can't use named aggregation with resample.

[alexjacobson95]

Code Sample, a copy-pastable example if possible

import pandas as pd
import numpy as np
from functools import partial

df = pd.DataFrame(
    np.random.randn(1000, 3), 
    index=pd.date_range('1/1/2012', freq='S', periods=1000), 
    columns=['A', 'B', 'C']
)
dfg = df.resample('3T').agg(
    {'A': [
        partial(np.quantile, q=.9999), 
        partial(np.quantile, q=.90),
    ]}
)

Problem description

Resample, unlike groupby, has no ability to do named aggregation. This means if you use a function twice on the same column you get an error saying:

pandas.core.base.SpecificationError: Function names must be unique, found multiple named quantile.

It would be very useful to make the resample agg interface similar to the groupby agg interface. An example of the named aggregation is here: https://pandas.pydata.org/pandas-docs/stable/user_guide/groupby.html#named-aggregation

Expected Output

Output of pd.show_versions()

INSTALLED VERSIONS ------------------ commit : None python : 3.6.9.final.0 python-bits : 64 OS : Darwin OS-release : 18.7.0 machine : x86_64 processor : i386 byteorder : little LC_ALL : None LANG : None LOCALE : en_US.UTF-8 pandas : 0.25.3 numpy : 1.17.4 pytz : 2019.3 dateutil : 2.8.1 pip : 18.1 setuptools : 40.6.2 Cython : None pytest : None hypothesis : None sphinx : None blosc : None feather : None xlsxwriter : None lxml.etree : None html5lib : None pymysql : None psycopg2 : None jinja2 : 2.10.3 IPython : None pandas_datareader: None bs4 : None bottleneck : None fastparquet : None gcsfs : None lxml.etree : None matplotlib : None numexpr : None odfpy : None openpyxl : None pandas_gbq : None pyarrow : 0.15.1 pytables : None s3fs : None scipy : None sqlalchemy : None tables : None xarray : None xlrd : None xlwt : None xlsxwriter : None

[TomAugspurger]

xref #28380

[charlesdong1991]

take

[MarcoGorelli]

Tried reproducing this on master - didn't get an error, but now the two aggregations are the same:

import pandas as pd
import numpy as np
from functools import partial

df = pd.DataFrame(
    np.random.randn(1000, 3), 
    index=pd.date_range('1/1/2012', freq='S', periods=1000), 
    columns=['A', 'B', 'C']
)
dfg = df.resample('3T').agg(
    {'A': [
        partial(np.quantile, q=.9999), 
        partial(np.quantile, q=.1111),
    ]}
)

returns

                            A          
                     quantile  quantile
2012-01-01 00:00:00 -1.035997 -1.035997
2012-01-01 00:03:00 -1.326052 -1.326052
2012-01-01 00:06:00 -1.267661 -1.267661
2012-01-01 00:09:00 -1.321961 -1.321961
2012-01-01 00:12:00 -1.103027 -1.103027
2012-01-01 00:15:00 -0.987614 -0.987614

[charlesdong1991]

thanks @MarcoGorelli I think you would need to give a name for this function since those two can not be distinguished for now, so it always takes the last one to use, code below seems to work on master

quant9 = partial(np.percentile, q=.9999)
quant1 = partial(np.percentile, q=.1111)
quant9.__name__ = "quant9"
quant1.__name__ = "quant1"

dfg = df.resample('3T').agg(
    {'A': [quant9, quant1,]}
)

assigning different names for it should work.

and also i haven't had time to really work on this recently, not sure how the result will go.

[MarcoGorelli]

i haven't had time to really work on this recently

Is this something you plan on coming back to? If so, that's fine, I'll leave it to you :) If not, I might have a solution in mind which doesn't involve renaming the functions.

[charlesdong1991]

awesome! pls go for it! @MarcoGorelli
i unassigned myself

[charlesdong1991]

btw, i think you might need to open a new issue and address this, because this seems not related to resample, but quite general groupby.agg @MarcoGorelli since currently implementation does not distinguish the example you provided.