DataFrame Aggregation and Grouping

Pandas aggregate() function is used to apply some aggregation across one or more column.

 
df.aggregate(['sum', 'min'])

Above statement will apply aggregation across all the columns in a DataFrame and calculate sum and min will be found for each numeric type column.

Lets create a DataFrame..

 
import pandas as pd
import numpy as np
df = pd.DataFrame()
df['Name'] = ['John', 'Doe', 'Bill','Jim','Harry','Ben']
df['Age'] = [14, 12, 14,11,12,14]
df['Category'] = ['A', 'E', 'B','E','C','D']
df['Height'] = [145, 152,167,136,149,161]
df['Weight'] = [34, 54,38,39,44,51]
df

 
    Name  Age Category  Height  Weight
0   John   14        A     145      34
1    Doe   12        E     152      54
2   Bill   14        B     167      38
3    Jim   11        E     136      39
4  Harry   12        C     149      44
5    Ben   14        D     161      51

The following operation will apply aggregation across all the columns in a DataFrame and calculate minimum and maximum will be found for each numeric type column.

 
df.aggregate(['min', 'max'])

 
     Name  Age Category  Height  Weight
min   Ben   11        A     136      34
max  John   14        E     167      54

DataFrame aggregation across different columns

 
df.aggregate({"Age":['min', 'max'],
              "Height":['mean', 'max'],
              "Weight":['mean', 'min']})

 
       Age      Height     Weight
max   14.0  167.000000        NaN
mean   NaN  151.666667  43.333333
min   11.0         NaN  34.000000

Aggregation with groupby

 
df.groupby('Age').agg({"Height": "sum"})

 
     Height
Age
11      136
12      301
14      473

For multiple functions applied for one column use a list of tuples - names of new columns and aggregated functions:

 
df.groupby(['Age'])['Height'].agg([('Average','mean'),('Total','sum')])

 
        Average  Total
Age
11   136.000000    136
12   150.500000    301
14   157.666667    473

If you want to pass multiple functions is possible pass list of tuples.

 
df.groupby(['Age', 'Category']).agg([('Average','mean'),('Total','sum')])

 
              Height        Weight
             Average Total Average Total
Age Category
11  E            136   136      39    39
12  C            149   149      44    44
    E            152   152      54    54
14  A            145   145      34    34
    B            167   167      38    38
    D            161   161      51    51

Instead of an aggregation function it is possible to pass list, tuple, set for converting column.

 
df.groupby('Age')['Category'].agg(list).reset_index()

 
   Age   Category
0   11        [E]
1   12     [E, C]
2   14  [A, B, D]

For converting to strings with separator use .join only if string column.

 
df.groupby('Age')['Category'].agg(','.join).reset_index()

 
   Age Category
0   11        E
1   12      E,C
2   14    A,B,D

Some common aggregating functions are tabulated below:

Function	Description
mean()	Compute mean of groups
sum()	Compute sum of group values
size()	Compute group sizes
count()	Compute count of group
std()	Standard deviation of groups
first()	Compute first of group values
last()	Compute last of group values
min()	Compute min of group values
max()	Compute max of group values

Next : How to Sort Pandas DataFrame