Pandas: Combine TimeGrouper with another Groupby argument

Question

I have the following DataFrame:

df = pd.DataFrame({
\'Branch\' : \'A A A A A B\'.split(),
\'Buyer\': \'Carl Mark Carl Joe Joe Carl\'.split(),
\'Quantity\': [         


        

Answer 1

You can now use a TimeGrouper with another column (as of IIRC pandas version 0.14):

In [11]: df1 = df.set_index('Date')

In [12]: g = df1.groupby([pd.TimeGrouper('20D'), 'Branch'])

In [13]: g.sum()
Out[13]:
                            Quantity
Date                Branch
2013-01-01 13:00:00 A              4
2013-09-18 13:00:00 A             13
2013-11-17 13:00:00 A              9
                    B              3

Answer 2

From the discussion here: https://github.com/pydata/pandas/issues/3791

In [38]: df.set_index('Date').groupby(pd.TimeGrouper('6M')).apply(lambda x: x.groupby('Branch').sum())
Out[38]: 
                   Quantity
           Branch          
2013-01-31 A              4
2014-01-31 A             22
           B              3

And a bit more complicated question

In [55]: def testf(df):
   ....:     if (df['Buyer'] == 'Mark').sum() > 0:
   ....:         return Series(dict(quantity = df['Quantity'].sum(), buyer = 'mark'))
   ....:     return Series(dict(quantity = df['Quantity'].sum()*100, buyer = 'other'))
   ....: 

In [56]: df.set_index('Date').groupby(pd.TimeGrouper('6M')).apply(lambda x: x.groupby('Branch').apply(testf))
Out[56]: 
                   buyer quantity
           Branch                
2013-01-31 A        mark        4
2014-01-31 A       other     2200
           B       other      300