Pandas Tutorial #14 – Sorting DataFrame

This tutorial will discuss different ways to sort a DataFrame row or column-wise.

Table Of Contents

• Sort all rows of DataFrame by a column
• Sort all rows of DataFrame by a column in Descending Order
• Sort DataFrame by row index labels
• Sort DataFrame by column names

First of all, we will create a DataFrame from a list of tuples,

import pandas as pd

# List of Tuples
empoyees = [(11, 'Jack',    44, 'Sydney',   19) ,
            (12, 'Riti',    41, 'Delhi' ,   17) ,
            (13, 'Aadi',    46, 'New York', 11) ,
            (14, 'Mohit',   45, 'Delhi' ,   15) ,
            (15, 'Veena',   43, 'Delhi' ,   14) ,
            (16, 'Shaunak', 42, 'Mumbai',   10 ),
            (17, 'Shaun',   40, 'Colombo',  12)]

# Create a DataFrame object
df = pd.DataFrame(  empoyees,
                    columns=['ID', 'Name', 'Age', 'City', 'Experience'],
                    index=['b', 'd', 'a', 'c', 'g', 'f', 'e'])

# Display the DataFrame
print(df)

Output:

   ID     Name  Age      City  Experience
b  11     Jack   44    Sydney          19
d  12     Riti   41     Delhi          17
a  13     Aadi   46  New York          11
c  14    Mohit   45     Delhi          15
g  15    Veena   43     Delhi          14
f  16  Shaunak   42    Mumbai          10
e  17    Shaun   40   Colombo          12

This DataFrame has seven rows and five columns. Now let’s see how we can sort this DataFrame based on its values or labels.

Sort all rows of DataFrame by a column

In Pandas, the DataFrame provides a method sort_values(), and it sorts the DataFrame by values along the given axis. We can sort the above created DataFrame by column ‘Experience’. For this we need to pass the name of this column as a list of columns in the by parameter of sort_values() i.e df.sort_values(by=[‘Experience’]) . It will sort all the rows in DataFrame by the column ‘Experience’. For example,

# Sort DataFrame by column 'Experience'
df = df.sort_values(by=['Experience'])

# Display the DataFrame
print(df)

Output:

   ID     Name  Age      City  Experience
f  16  Shaunak   42    Mumbai          10
a  13     Aadi   46  New York          11
e  17    Shaun   40   Colombo          12
g  15    Veena   43     Delhi          14
c  14    Mohit   45     Delhi          15
d  12     Riti   41     Delhi          17
b  11     Jack   44    Sydney          19

It sorted the DataFrame along the ‘index’ axis, i.e., sorted all the rows along the column ‘Experience’. In this example, we sorted the DataFrame along a numeric column. We can also sort a DataFrame along a string column. For example,

# Sort DataFrame by column 'Name'
df = df.sort_values(by=['Name'])

# Display the DataFrame
print(df)

Output:

   ID     Name  Age      City  Experience
a  13     Aadi   46  New York          11
b  11     Jack   44    Sydney          19
c  14    Mohit   45     Delhi          15
d  12     Riti   41     Delhi          17
e  17    Shaun   40   Colombo          12
f  16  Shaunak   42    Mumbai          10
g  15    Veena   43     Delhi          14

It sorted the DataFrame along the column ‘Name’. This column contains the string values; therefore, the set_value() method sorted the rows of DataFrame based on the alphabetical order of column ‘Name’ values.

Sort all rows of DataFrame by a column in Descending Order

To sort the DataFrame in descending order, pass the argument ascending=False in the sort_values() function. For example,

# Sort DataFrame by column 'Experience' in descending order
df = df.sort_values(by=['Experience'], ascending=False)

# Display the DataFrame
print(df)

Output:

   ID     Name  Age      City  Experience
b  11     Jack   44    Sydney          19
d  12     Riti   41     Delhi          17
c  14    Mohit   45     Delhi          15
g  15    Veena   43     Delhi          14
e  17    Shaun   40   Colombo          12
a  13     Aadi   46  New York          11
f  16  Shaunak   42    Mumbai          10

It sorted all the rows of DataFrame along the column ‘Experience’ in descending order.

Sort DataFrame by row index labels

In Pandas, the DataFrame provides a method sort_index(), and it sorts the DataFrame by index labels along the given axis. By default, it sorts the rows of DataFrame based on row index labels. For example,

# Sort DataFrame by the Row Index labels
df = df.sort_index()

# Display the DataFrame
print(df)

Output:

   ID     Name  Age      City  Experience
a  13     Aadi   46  New York          11
b  11     Jack   44    Sydney          19
c  14    Mohit   45     Delhi          15
d  12     Riti   41     Delhi          17
e  17    Shaun   40   Colombo          12
f  16  Shaunak   42    Mumbai          10
g  15    Veena   43     Delhi          14

It sorted all the rows of DataFrame by the row index labels.

Sort DataFrame by column names

Pass the axis=1 argument in the sort_index() method of DataFrame. It will sort the DataFrame by the column names. For example,

# Sort DataFrame by the Column Names
df = df.sort_index(axis=1)

# Display the DataFrame
print(df)

Output:

   Age      City  Experience  ID     Name
b   44    Sydney          19  11     Jack
d   41     Delhi          17  12     Riti
a   46  New York          11  13     Aadi
c   45     Delhi          15  14    Mohit
g   43     Delhi          14  15    Veena
f   42    Mumbai          10  16  Shaunak
e   40   Colombo          12  17    Shaun

It sorted all the columns of DataFrame by the column names.

Summary

We learned about different ways to sort a DataFrame in Pandas.