Python pandas apply function to column – pandas.apply(): Apply a function to each row/column in Dataframe

by

How to apply a function  to each row or column in Dataframe in Python.

Python pandas apply function to column: To apply a function to each row or column data in a warframe be it lambda, user-defined or a numpy function we have to use a function from Python’s Pandas library.  The function belongs to the dataframe class .

Syntax-

DataFrame.apply(func, axis=0, broadcast=None, raw=False, reduce=None, result_type=None, args=(), **kwds)

Arguments :

  1. Func : It is the function that is to be applied to the rows/columns. It takes series as arguments and also returns series
  2. Axis : Axis is the axis in which the function is applied to the rows/columns. (default value is 0. If 1 means it applies to all rows, If 0 means it applies to all columns)
  3. Args : All the arguments passed in a list of tuples.

Apply a lambda function to each row or each column in Dataframe :

Let us consider a lambda function

lambda x : x + 10

Apply a lambda function to each column :

To apply the function to each column we just have to pass the lambda function as argument in the Dataframe.apply( ) function.

#Program :

import pandas as pd
import numpy as np
#list of tuples
matrix = [(222, 34, 23),
         (333, 31, 11),
         (444, 16, 21),
         (555, 32, 22),
         (666, 33, 27),
         (777, 35, 11)
         ]
#creating an object from Dataframe class
dfObj = pd.DataFrame(matrix, columns=list('abc'))
#Passing only the lambda function into the Dataframe function so that it gets applied to columns only
modMatrix = dfObj.apply(lambda x : x + 10)
print("After applying the lambda function to each column in dataframe")
print(modMatrix)

Output :

After applying the lambda function to each column in dataframe

     a      b   c
0  232  44  33
1  343  41  21
2  454  26  31
3  565  42  32
4  676  43  37
5  787  45  21

Apply a lambda function to each row :

To apply the function to each row we just have to add axis=1 and pass it to the lambda function with the lambda function in the Dataframe.apply( ) function like we did in the column.

#Program :

import pandas as pd
import numpy as np
#list of tuples
matrix = [(222, 34, 23),
         (333, 31, 11),
         (444, 16, 21),
         (555, 32, 22),
         (666, 33, 27),
         (777, 35, 11)
         ]
#creating an object from Dataframe class
dfObj = pd.DataFrame(matrix, columns=list('abc'))
#Passing only the lambda function into the Dataframe function so that it gets applied to columns only
modMatrix = dfObj.apply(lambda x : x + 10)
print("After applying the lambda function to each row in dataframe")
print(modMatrix)

Output :
After applying the lambda function to each row in dataframe
     a      b    c
0  227  39  28
1  338  36  16
2  449  21  26
3  560  37  27
4  671  38  32
5  782  40  16

Apply a User Defined function with or without arguments to each row or column of a Dataframe :

For this let us consider a user-defined function that multiplies the values by 2

def doubleData(x):



 return x * 2


Apply a user-defined function to each column :


Like we applied the lambda function to each column, similarly we will only pass the function here.


#Program :

import pandas as pd
import numpy as np
# Multiply given value by 2 and returns
def doubleData(x):
   return x * 2
#list of tuples
matrix = [(222, 34, 23),
         (333, 31, 11),
         (444, 16, 21),
         (555, 32, 22),
         (666, 33, 27),
         (777, 35, 11)
         ]
#creating an object from Dataframe class
dfObj = pd.DataFrame(matrix, columns=list('abc'))
#Applyin the user defined function doubleData to columns only
modMatrix = dfObj.apply(doubleData)
print("After applying the user-defined function to each column in dataframe")
print(modMatrix)



Output :
After applying the user-defined function to each column in dataframe
      a      b    c
0   444  68  46
1   666  62  22
2   888  32  42
3  1110  64  44
4  1332  66  54
5  1554  70  22


Apply a user-defined function to each row :


We just have to add axis=1 to the above function to apply it to rows.


#Program :

import pandas as pd
import numpy as np
# Multiply given value by 2 and returns
def doubleData(x):
   return x * 2
#list of tuples
matrix = [(222, 34, 23),
         (333, 31, 11),
         (444, 16, 21),
         (555, 32, 22),
         (666, 33, 27),
         (777, 35, 11)
         ]
#creating an object from Dataframe class
dfObj = pd.DataFrame(matrix, columns=list('abc'))
#Applyin the user defined function doubleData to rows only
modMatrix = dfObj.apply(doubleData,axis=1)
print("After applying the user-defined function to each row in dataframe")
print(modMatrix)



Output :
After applying the user-defined function to each row in dataframe
      a      b    c
0   444  68  46
1   666  62  22
2   888  32  42
3  1110  64  44
4  1332  66  54
5  1554  70  22





Apply a user-defined function to each row or column with arguments :


Let us take a user defined function that takes accepts a series and a number, then returns the series multiplied to the number


#Program :

import pandas as pd
import numpy as np
#Multplies the whole seried with the number and return the series
def multiplyData(x, y):
   return x * y
#list of tuples
matrix = [(222, 34, 23),
         (333, 31, 11),
         (444, 16, 21),
         (555, 32, 22),
         (666, 33, 27),
         (777, 35, 11)
         ]
#creating an object from Dataframe class
dfObj = pd.DataFrame(matrix, columns=list('abc'))
#Applyin the user defined function with a argument
modMatrix = dfObj.apply(multiplyData, args=[4])
print("After applying the user-defined function with argument in dataframe")
print(modMatrix)



Output :
After applying the user-defined function with argument in dataframe
      a       b      c
0   888  136   92
1  1332  124   44
2  1776   64   84
3  2220  128   88
4  2664  132  108
5  3108  140   44


Apply a numpy functions to each row or column of a Dataframe


For this let’s use the numpy function numpy.square( ). (For columns pass the function directly and for rows add axis=1 and pass)


#Program

import pandas as pd
import numpy as np
#list of tuples
matrix = [(222, 34, 23),
         (333, 31, 11),
         (444, 16, 21),
         (555, 32, 22),
         (666, 33, 27),
         (777, 35, 11)
         ]
#creating an object from Dataframe class
dfObj = pd.DataFrame(matrix, columns=list('abc'))
#Applyin the numpy fuction .square()
modMatrix = dfObj.apply(np.square)
print("After applying the numpy function in dataframe")
print(modMatrix)



Output :
After applying the numpy function in dataframe
        a           b     c
0   49284  1156  529
1  110889   961  121
2  197136   256  441
3  308025  1024  484
4  443556  1089  729
5  603729  1225  121


Apply a Reducing functions to a to each row or column of a Dataframe


We passed a series into the user-defined functions and it also returned a series .  However, we can also pass a series and return a single variable. Let’s use numpy.sum( ) for that.


#Program :

import pandas as pd
import numpy as np
#list of tuples
matrix = [(222, 34, 23),
         (333, 31, 11),
         (444, 16, 21),
         (555, 32, 22),
         (666, 33, 27),
         (777, 35, 11)
         ]
#creating an object from Dataframe class
dfObj = pd.DataFrame(matrix, columns=list('abc'))
#Applyin the numpy fuction .sum()
modMatrix = dfObj.apply(np.sum)
print("After applying the numpy function in datframe")
print(modMatrix)



Output :
After applying the numpy function in datframe
a    2997
b     181
c     115
dtype: int64


Want to expert in the python programming language? Exploring Python Data Analysis using Pandas tutorial changes your knowledge from basic to advance level in python concepts.


Read more Articles on Python Data Analysis Using Padas – Modify a Dataframe