Apply a function to each row or column in Dataframe using pandas.apply()

There are different ways to apply a function to each row or column in DataFrame. We will learn about various ways in this post. Let’s create a small dataframe first and see that. 

Python3

# import pandas and numpy library
import pandas as pd
import numpy as np

# list of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Create a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Output
df

Output :

Method 1: Applying lambda function to each row/column. 
Example 1: For Column 

Python3

# import pandas and numpy library
import pandas as pd
import numpy as np

# list of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Create a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a lambda function to each
# column which will add 10 to the value
new_df = df.apply(lambda x : x + 10)

# Output
new_df

Output :

Example 2: For Row 

Python3

# import pandas and numpy library
import pandas as pd
import numpy as np

# list of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a lambda function to each
# row which will add 5 to the value
new_df = df.apply(lambda x: x + 5, axis = 1)

# Output
new_df

Output :

Method 2: Applying user defined function to each row/column 
Example 1: For Column  

Python3

# function to returns x*x
def squareData(x):
    return x * x

# import pandas and numpy packages
import pandas as pd
import numpy as np

# list of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a user defined function to
# each column that will square the given
# value
new_df = df.apply(squareData)

# Output
new_df

Output :

Example 2: For Row  

Python3

# function to returns x*X
def squareData(x):
    return x * x

# import pandas and numpy library
import pandas as pd
import numpy as np

# List of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a user defined function
# to each row that will square the given value
new_df = df.apply(squareData, axis = 1)

# Output
new_df

Output :

In the above examples, we saw how a user defined function is applied to each row and column. We can also apply user defined functions which take two arguments. 

Example 1: For Column 

Python3

# function to returns x+y
def addData(x, y):
    return x + y

# import pandas and numpy library
import pandas as pd
import numpy as np

# list of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a user defined function to each
# column which will add value in each
# column by given number
new_df = df.apply(addData, args = [1])

# Output
print(new_df)

Output:

Example 2: For Row 

Python3

# function to returns x+y
def addData(x, y):
    return x + y

# import pandas and numpy library
import pandas as pd
import numpy as np

# List of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a user defined function to each
# row which will add value in each row by
# given number
new_df = df.apply(addData, axis = 1,
                    args = [3])

# Output
new_df

Output :

Method 3: Applying numpy function to each row/column 
Example 1: For Column 

Python3

# import pandas and numpy library
import pandas as pd
import numpy as np

# list of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a numpy function to each
# column by squaring each value
new_df = df.apply(np.square)

# Output
new_df

Output :

Example 2: For Row 

Python3

# import pandas and numpy library
import pandas as pd
import numpy as np

# List of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Apply a numpy function to each row
# to find square root of each value
new_df = df.apply(np.sqrt, axis = 1)

# Output
new_df

Output :

Method 4: Applying a Reducing function to each row/column 
A Reducing function will take row or column as series and returns either a series of same size as that of input row/column or it will return a single variable depending upon the function we use.

Example 1: For Column 

Python3

# import pandas and numpy library
import pandas as pd
import numpy as np

# List of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a numpy function to get the sum
# of all values in each column
new_df = df.apply(np.sum)

# Output
new_df

Output :

Example 2: For Row 

Python3

# import pandas and numpy library
import pandas as pd
import numpy as np

# List of tuples
matrix = [(1,2,3,4),
          (5,6,7,8,),
          (9,10,11,12),
          (13,14,15,16)
         ]

# Creating a Dataframe object
df = pd.DataFrame(matrix, columns = list('abcd'))

# Applying a numpy function to get t
# he sum of all values in each row
new_df = df.apply(np.sum, axis = 1)

# Output
new_df

Output :