The DataFrame reset_index() method in Pandas resets the index of a DataFrame, replacing it with the default integer index (starting from 0). If the DataFrame has a MultiIndex, it can selectively reset one or more levels.

import pandas as pd

df = pd.DataFrame(
    {
        'Team_Name': ["Lakers", "Patriots", "Yankees", "Lakers", "Red Sox", "Warriors", "Patriots"],
        'Wins': [12, 10, 14, 12, 13, 15, 18]
    },
    index=[101, 102, 103, 104, 105, 106, 107]
)

print(df)

print(df.reset_index())

 

In this code, you can see that, first, we had an index of [101, 102, 103, 104, 105, 106, 107] while creating a DataFrame.

Then, we reset the index using the df.reset_index() method. Now, that “index” has become a new column, and the new index will be: [0, 1, 2, 3, 4, 5, 6].

Syntax

DataFrame.reset_index(level=None, 
                      drop=False, 
                      inplace=False, 
                      col_level=0, 
                      col_fill="",
                      allow_duplicates=lib.no_default,
                      names=None)

Parameters

Argument	Description
level	It specifies the level(s) to reset in a MultiIndex.
drop	If it is True, it discards the original index instead of inserting it as columns.
inplace	If it is True, it changes the DataFrame in place and returns None.
col_level	If your DataFrame contains MultiIndex columns, it specifies the level where the reset index labels are inserted.
col_fill	For MultiIndex columns, this argument value fills in other levels when inserting the index.  If None, repeats the index name.
allow_duplicates	It allows duplicate column labels when inserting the index.
names	It renames the column(s) created from the reset index.

Using drop=True to discard the original Index

The drop=True argument is essential when you are resetting the index. But the question is, why? Well, if you are resetting the index, the old index will become a new column, and the latest index, starting from 0, will be inserted.

So, we can remove the old index column by passing the argument drop=True. It makes it a clean DataFrane with only a single index.

It basically prevents inserting the index as a column.

import pandas as pd

df = pd.DataFrame(
    {
        'Team_Name': ["Lakers", "Patriots", "Yankees", "Lakers", "Red Sox", "Warriors", "Patriots"],
        'Wins': [12, 10, 14, 12, 13, 15, 18]
    },
    index=[101, 102, 103, 104, 105, 106, 107]
)

print(df)

reset_df = df.reset_index(drop=True)

print(reset_df)

Handling multiIndex

Let’s make “Team_Name” and “Wins” into a MultiIndex using the set_index() method while adding a new column “Loses”.

import pandas as pd

df = pd.DataFrame(
    {
        'Team_Name': ["Lakers", "Patriots", "Yankees", "Lakers", "Red Sox", "Warriors", "Patriots"],
        'Wins': [12, 10, 14, 12, 13, 15, 18],
        'Losses': [5, 7, 6, 4, 8, 3, 9]
    },
    index=[101, 102, 103, 104, 105, 106, 107]
)

print(df)

df_multi = df.set_index(["Team_Name", "Wins"])

print(df_multi)

 

Reset the entire multiIndex

Let’s reset the entire multiindex by using the df.reset_index() method.

So, the new DataFrame will have an index that starts at 0, and the old two indexes will become the columns of the DataFrame.

import pandas as pd

df = pd.DataFrame(
    {
        'Team_Name': ["Lakers", "Patriots", "Yankees", "Lakers", "Red Sox", "Warriors", "Patriots"],
        'Wins': [12, 10, 14, 12, 13, 15, 18],
        'Losses': [5, 7, 6, 4, 8, 3, 9]
    },
    index=[101, 102, 103, 104, 105, 106, 107]
)

print(df)

df_multi = df.set_index(["Team_Name", "Wins"])

print(df_multi)

reset_df = df_multi.reset_index()

print(reset_df)

The above output figure shows that both index levels have returned to their respective columns.

Reset only one level (e.g., Team_Name)

Let’s reset only the “Team_Name” column so that it will become a regular column, and “Wins” will still be an “index”.

import pandas as pd

df = pd.DataFrame(
    {
        'Team_Name': ["Lakers", "Patriots", "Yankees", "Lakers", "Red Sox", "Warriors", "Patriots"],
        'Wins': [12, 10, 14, 12, 13, 15, 18],
        'Losses': [5, 7, 6, 4, 8, 3, 9]
    },
    index=[101, 102, 103, 104, 105, 106, 107]
)

print(df)

df_multi = df.set_index(["Team_Name", "Wins"])

print(df_multi)

reset_one_level_df = df_multi.reset_index(level="Team_Name")

print(reset_one_level_df)

 

That’s all!