Pandas DataFrame.reset_index()

The reset_index() method in Pandas is used to manage and reset the index of a DataFrame. It is useful after performing operations that modify the index such as filtering, grouping or setting a custom index. By default reset_index() reverts to a clean, default integer-based index (0, 1, 2, ...) which makes the DataFrame easier to work with.

Now lets see a basic example:

import pandas as pd

data = {'Name': ['Alice', 'Bob', 'Max'], 'Age': [25, 30, 35]}
df = pd.DataFrame(data, index=['a', 'b', 'c'])

print("Original DataFrame:")
print(df)

df_reset = df.reset_index()

print("\nDataFrame after reset_index():")
print(df_reset)

Output:

This code creates a DataFrame with 'Name' and 'Age' columns and a custom index ('a', 'b', 'c'). The reset_index() function moves these index labels into a new 'index' column and replaces them with default row numbers (0, 1, 2).

Syntax

DataFrame.reset_index(level=None, drop=False, inplace=False, col_level=0, col_fill='')

Parameters:

• level (optional): Specifies the index level to reset (useful for multi-level indices). It can be an integer, string or a list of levels.
• drop (default: False): If True, removes the old index instead of adding it as a column.
• inplace (default: False): If True, modifies the original DataFrame in place. If False, returns a new DataFrame.
• col_level (default: 0): Used to select the level of the column to insert the index labels.
• col_fill (default: ''): If the DataFrame has multiple column levels, this determines how missing levels are filled in the column headers.

Returns: The reset_index() method returns a new DataFrame with the index reset, unless inplace=True is specified. In that case, the original DataFrame is modified directly without creating a new one.

Lets see some other examples for DataFrame.reset_index() to understand it in a better way.

Example 1: Resetting Index of a Pandas DataFrame

In this example, we will set the "First Name" column as the index of the DataFrame and then reset the index using the reset_index() method. This process will move the custom index back to a regular column and restore the default integer-based index.

import pandas as pd

data = pd.DataFrame({
    'First Name': ['John', 'Jane', 'Emily', 'Michael', 'Sara'],
    'Last Name': ['Doe', 'Smith', 'Jones', 'Brown', 'Davis'],
    'Age': [28, 34, 22, 45, 29],
    'Department': ['HR', 'Finance', 'IT', 'Marketing', 'Sales'],
    'Salary': [55000, 68000, 48000, 75000, 60000]
})

print("Original Employee Dataset:")
display(data.head())

data.set_index("First Name", inplace=True)
print("\nAfter Setting 'First Name' as Index:")
display(data.head())

data.reset_index(inplace=True)
print("\nAfter Resetting Index:")
display(data.head())

Output:

Example 2: Resetting the Index After Filtering Data

When filtering a DataFrame, the original row indices are retained which can lead to inconsistencies during further operations. Using reset_index() ensures that the index is clean and sequential.

import pandas as pd

data = {'ID': [101, 102, 103, 104, 105],
        'Name': ['Alice', 'Bob', 'Charlie', 'David', 'Emma'],
        'Dept': ['HR', 'IT', 'IT', 'Finance', 'HR'],
        'Salary': [50000, 60000, 65000, 70000, 55000]}

df = pd.DataFrame(data)
print("Original DataFrame:")
print(df)

df_it = df[df['Dept'] == 'IT']
print("\nFiltered DataFrame:")
print(df_it)

df_it = df_it.reset_index(drop=True)
print("\nFiltered DataFrame after reset_index():")
print(df_it)

Output:

The filtered DataFrame (df_filtered) contains only employees from the IT department but retains the original row indices (1 and 2). By applying reset_index(drop=True), we remove the old indices and replace them with a clean, default integer index (0, 1) which results in a more structured and easier-to-analyze DataFrame.

Example 3: Resetting Index for Multi-Level DataFrames

If our DataFrame has a multi-level index, reset_index() can reset one or more of the index levels, turning them back into regular columns.

import pandas as pd

data = {'Region': ['North', 'North', 'South', 'South'],
        'City': ['New York', 'Los Angeles', 'Miami', 'Houston'],
        'Population': [8.4, 3.9, 0.4, 2.3]}
df = pd.DataFrame(data)

df.set_index(['Region', 'City'], inplace=True)

print("DataFrame with Multi-Level Index:")
print(df)

df_reset = df.reset_index(level='City')

print("\nDataFrame after resetting the 'City' level of the index:")
print(df_reset)

Output:

Key Use Cases of reset_index()

1. Simplifying Data Manipulation: After performing operations like filtering or sorting, we may want to reset the index to have a clean, sequential index.
2. Handling Multi-Level Indexes: When working with multi-level indexes, it can be used to remove specific index levels without affecting the rest of the structure.
3. Restoring Default Index: If you've set a custom index using the set_index() method, reset_index() restores the default integer-based index.

By understanding and using reset_index(), we can efficiently manage and reorganize our DataFrame's index which makes our data easier to manipulate and analyze in various situations.