How to reset index after Groupby pandas?

Last Updated : 11 Dec, 2020

Python’s groupby() function is versatile. It is used to split the data into groups based on some criteria like mean, median, value_counts, etc. In order to reset the index after groupby() we will use the reset_index() function.

Below are various examples which depict how to reset index after groupby() in pandas:

Example 1

Python3

# import required modules 
import numpy as np 
import pandas as pd 
  
# creating dataframe 
df = pd.DataFrame({'Subject': ['Physics',  
                               'Chemistry',  
                               'Maths'],  
                   'Marks': [4, 8, 5]}) 
  
# grouping the data on the basis of  
# subject and mean of marks. 
df_grouped = df.groupby(['Subject']).mean() 
  
# display dataset 
df_grouped

Output:

Resetting the index after grouping data, using reset_index(), it is a function provided by python to add indexes to the data.

Python3

# reset index 
df_grouped.reset_index()

Output:

Example 2:

Creating Dataframe.

Python3

# import required modules 
import pandas as pd 
import numpy as np 
  
# creating dataframe 
df2 = pd.DataFrame({'Student': [1, 2, 3, 4, 1, 3, 2, 4, 1, 2, 4, 3], 
                    'Amount': [ 
                   10, 20, 30, 40, 20, 60, 40, 80, 30, 60, 120, 90]}) 
  
# grouping the data 
df2_group = df2.groupby(['Student']) 
  
# grouped on the basis of students and 
# with the value of count of amount 
df2_grouped = df2_group['Amount'].value_counts() 
  
# display dataset 
print(df2_grouped) 

Output:

Resetting the index. This will give you an error.

Python3

# this will generate an error. 
df2_grouped.reset_index() 

Output:

Naming the reset_index() will group and reset the index.

Python3

# resetting index on the basis of count 
df2_grouped.reset_index(name = 'count')

Output:

Example 3

Here, is another example to depict how to reset dataframe after using groupby().

Python3

# import required modules 
import numpy as np 
import pandas as pd 
  
# creating dataframe 
df = pd.DataFrame({'Subject': ['B',  
                               'C',  
                               'A','D','C','B','A'],  
                   'Marks': [4, 8, 5,9,8,1,0]}) 
  
# grouping the data on the basis of  
# subject and mean of marks. 
df_grouped = df.groupby(['Subject']).mean() 
  
# display dataset 
df_grouped 
  
# reset index 
df_grouped.reset_index()