Pandas apply() Function to Single & Multiple Column(s)

Using Pandas.DataFrame.apply() method you can execute a function to a single column, all, and a list of multiple columns (two or more). In this article, I will cover how to apply() function on values of a selected single, multiple, and all columns. For example, let's say we have three columns and would like to apply a function on a single column without touching the other two columns and return a DataFrame with three columns.

Key Points –

Pandas' apply() function is a powerful tool for applying a function along one or more axes of a DataFrame.
When applied to a single column, apply() iterates over each element of the column, applying the specified function.
For multiple columns, apply() can operate on either rows or columns, based on the axis parameter.
The apply() function can significantly enhance the efficiency of data manipulation tasks by allowing custom operations on DataFrame elements

Quick Examples of apply() Function to Single & Multiple Column(s)

If you are in a hurry, below are some quick examples of how to apply a function to single and multiple columns (two or more) in Pandas DataFrame.


# Quick examples of pandas apply() function 

# Example 1: Using Dataframe.apply() 
# To apply function add column
def add_3(x):
   return x+3
df2 = df.apply(add_3)

# Example 2: Using apply function single column
def add_4(x):
   return x+4
df["B"] = df["B"].apply(add_4)

# Example 3: Apply to multiple columns
df[['A','B']] = df[['A','B']].apply(add_3)

# Example 4: Apply a lambda function to each column
df2 = df.apply(lambda x : x + 10)

# Example 5: Using Dataframe.apply() and lambda function
df["A"] = df["A"].apply(lambda x: x-2)

# Example 6: Using Dataframe.apply() & [] operator
df['A'] = df['A'].apply(np.square)

# Example 7: Using numpy.square() and [] operator
df['A'] = np.square(df['A'])

# Example 8: Apply function numpy.square() 
# To square the values of two rows 
#'A'and'B
df2 = df.apply(lambda x: np.square(x) if x.name in ['A','B'] else x)

# Example 9: Apply function single column 
# Using transform() 
def add_2(x):
    return x+2
df = df.transform(add_2)

# Example 10: Using DataFrame.map() 
# To Single Column
df['A'] = df['A'].map(lambda A: A/2.)

# Example 11: Using DataFrame.assign() and lambda
df2 = df.assign(B=lambda df: df.B/2)

Syntax of Pandas.DataFrame.apply() Function

If you are a learner let's see the syntax of the apply() method and execute some examples of how to apply it on a single column, multiple, and all columns.


# Syntax of apply() function
DataFrame.apply(func, axis=0, raw=False, result_type=None, args=(), **kwargs)

To run some examples of pandas applying functions to single & multiple column(s), let's create DataFrame with column names A, B, and C.


# Create DataFrame
import pandas as pd
import numpy as np
data = [(3,5,7), (2,4,6),(5,8,9)]
df = pd.DataFrame(data, columns = ['A','B','C'])
print("Create DataFrame:\n", df)

Yields below output.

Pandas Apply Function to Single Column

You can define a function add_4() that adds value 4 for every value in a specified column value and uses this on the apply() function. To apply it to a single column, qualify the column name using df["col_name"]. The example below applies a function to a column B.


# Using apply function single column
def add_4(x):
   return x+4
df["B"] = df["B"].apply(add_4)
print("After applying a function on a single column:\n", df)

Yields below output. This applies the function to every row in DataFrame for a specified column.

Pandas Apply Function to All Columns

In some cases we would want to apply a function on all pandas columns, you can do this using the apply() function. Here the add_3() function will be applied to all DataFrame columns.


# Using Dataframe.apply() to apply function add column
def add_3(x):
   return x+3
df2 = df.apply(add_3)
print("After applying a function on multiple columns:\n", df2)

# Output:
# After applying a function on multiple columns:
#   A   B   C
#0  6   8  10
#1  5   7   9
#2  8  11  12

Pandas Apply Function to Lists of Columns

Similarly, you can apply a function on a selected list of columns. In this case, this function will apply to only the selected list of columns without touching the rest of the columns.


# Apply() function on selected list of multiple columns
df = pd.DataFrame(data, columns = ['A','B','C'])
df[['A','B']] = df[['A','B']].apply(add_3)
print("After applying a function on list of columns:\n", df)

# Output:
# After applying a function on list of columns:
#   A   B  C
#0  6   8  7
#1  5   7  6
#2  8  11  9

Apply the Lambda Function to Each Column

You can also use a lambda expression with the apply() method to manipulate each column value in a DataFrame, the Below example, adds 10 to all column values.


# Apply a lambda function to each column
df2 = df.apply(lambda x : x + 10)
print("After applying a lambda function:\n", df2)

# Output:
# After applying a lambda function:
#    A   B   C
#0  13  15  17
#1  12  14  16
#2  15  18  19

Apply the Lambda Function to a Single Column

Applying a lambda function to a single column allows you to perform elelement-wise operations on that particular column. For instance, you applied the lambda function lambda x: x - 2 to the column A, subtracting 2 from each element in that column.


# Using Dataframe.apply() and lambda function
df["A"] = df["A"].apply(lambda x: x-2)
print("After applying a lambda function to a single column:\n", df)

# Output:
# After applying a lambda function to a single column:
#   A  B  C
#0  1  5  7
#1  0  4  6
#2  3  8  9

Using DataFrame.transform() to Apply() Function

Using DataFrame.apply() method & lambda functions the resultant DataFrame can be any number of columns whereas with transform() function the resulting DataFrame must have the same length as the input DataFrame.


# Using DataFrame.transform() 
def add_2(x):
    return x+2
df = df.transform(add_2)
print("After applying a function to every column:\n", df)

# Output:
# After applying a function to every column:
#   A   B   C
#0  5   7   9
#1  4   6   8
#2  7  10  11

Single Column Using DataFrame.map()

Here is another alternative using map() function to a single column in a DataFrame. For instance, the lambda function lambda x: x/2 is applied to each element in the column A, dividing each value by 2.0.


# Using DataFrame.map() to Single Column
df['A'] = df['A'].map(lambda A: A/2.)
print("After applying a function to a single column:\n", df)

# Output:
# After applying a function to a single column:
#     A  B  C
#0  1.5  5  7
#1  1.0  4  6
#2  2.5  8  9

Using DataFrame.assign() and Lambda Function

You can also try assign() with a lambda function to apply it to a single column in a DataFrame. For instance, the lambda function lambda x: x['B']/2 is applied within assign() to create a new DataFrame (df2) where the column B is modified by dividing each value by 2.0.


# Using DataFrame.assign() and Lambda
df2 = df.assign(B=lambda df: df.B/2)
print("After applying a function to a single column:\n", df2)

# Output:
# After applying a function to a single column:
#   A    B  C
#0  3  2.5  7
#1  2  2.0  6
#2  5  4.0  9

Using the Numpy Function on a Single Column

Use df['A']=df['A'].apply(np.square) to select the column from DataFrame as a series using the [] operator and apply numpy.square() method.


# Using Dataframe.apply() & [] operator
df['A'] = df['A'].apply(np.square)
print("After applying a function to a column:\n", df)

# Output:
# After applying a function to a column:
#    A  B  C
#0   9  5  7
#1   4  4  6
#2  25  8  9

Using NumPy.square() Method

You can use NumPy's square() method to square the values in a DataFrame column. For example, np.square(df['A']) square each element in column A, and the result is assigned back to column A in the DataFrame.


# Using numpy.square() and [] operator
df['A'] = np.square(df['A'])
print("After applying a function to a column:\n", df)

Yields the same output as above.

Using NumPy.square() & Lambda Function to Multiple Columns

Use apply() and a lambda function to conditionally square the values of columns A and B using NumPy's square() function. For instance, the lambda function checks if the name of the current column (x.name) is either A or B. If so, it applies np.square(x) to square the values in that column. Otherwise, it leaves the column unchanged.


# Apply function NumPy.square() to square the values of two rows 
# 'A'and'B'
df2 = df.apply(lambda x: np.square(x) if x.name in ['A','B'] else x)
print("After applying a lambda function to multiple columns:\n", df2)

# Output:
# After applying a lambda function to multiple columns:
#    A   B  C
#0   9  25  7
#1   4  16  6
#2  25  64  9

FAQ on Pandas apply() Function

apply() is a Pandas DataFrame method used to apply a function along the axis of a DataFrame. It can be used to perform operations on both rows and columns.

When applied to a single column, apply() applies the specified function to each element in that column. For instance, df['column'] = df['column'].apply(lambda x: x * 2)

you can use apply() on multiple columns by specifying the axis parameter. When applying to columns, use axis=0. For instance, df[['column1', 'column2']] = df[['column1', 'column2']].apply(lambda x: x * 2, axis=0)

You can use any custom function with apply(). Just define your function and pass it as an argument. For instance, def add_3(x): return x+3 df2 = df.apply(add_3)

Conclusion

In this article, you have learned how to apply a function to a single column, all, and multiple columns (two or more) of Pandas DataFrame using apply(), transform(), NumPy.square(), map(), transform(), and assign() methods.

Happy Learning !!

References

https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.transform.html

Pandas apply() Function to Single &amp; Multiple Column(s)