How to Sum Specific Rows in Pandas (With Examples)

You can use the following methods to find the sum of specific rows in a pandas DataFrame:

Method 1: Sum Specific Rows by Index

#sum rows in index positions 0, 1, and 4
df.iloc[[0, 1, 4]].sum()

Method 2: Sum Specific Rows by Label

#sum rows with index labels 'A', 'B', and 'E'
df.loc[['A', 'B', 'E']].sum()

The following examples show how to use each method in practice with the following pandas DataFrame:

import pandas as pd

#create DataFrame
df = pd.DataFrame({'points': [28, 17, 19, 14, 23, 26, 5],
                   'rebounds': [5, 6, 4, 7, 14, 12, 9],
                   'assists': [10, 13, 7, 8, 4, 5, 8]})

#set index
df = df.set_index([pd.Index(['A', 'B', 'C', 'D', 'E', 'F', 'G'])])

#view DataFrame
print(df)

   points  rebounds  assists
A      28         5       10
B      17         6       13
C      19         4        7
D      14         7        8
E      23        14        4
F      26        12        5
G       5         9        8

Example 1: Sum Specific Rows by Index

The following code shows how to sum the values in the rows with index values 0, 1, and 4 for each column in the DataFrame:

#sum rows in index positions 0, 1, and 4
df.iloc[[0, 1, 4]].sum()

points      68
rebounds    25
assists     27
dtype: int64

From the output we can see:

The sum of rows with index values 0, 1, and 4 for the points column is 68.
The sum of rows with index values 0, 1, and 4 for the rebounds column is 25.
The sum of rows with index values 0, 1, and 4 for the assists column is 27.

Also note that you can sum a specific range of rows by using the following syntax:

#sum rows in index positions between 0 and 4
df.iloc[0:4].sum()

points      78
rebounds    22
assists     38
dtype: int64

From the output we can see the sum of the rows with index values between 0 and 4 (not including 4) for each of the columns in the DataFrame.

Example 2: Sum Specific Rows by Label

The following code shows how to sum the values in the rows with index labels ‘A’, ‘B’, and ‘E’ for each column in the DataFrame:

#sum rows with index labels 'A', 'B', and 'E'
df.loc[['A', 'B', 'E']].sum()

points      68
rebounds    25
assists     27
dtype: int64

From the output we can see:

The sum of rows with index values ‘A’, ‘B’, and ‘E’ for the points column is 68.
The sum of rows with index values ‘A’, ‘B’, and ‘E’ for the rebounds column is 25.
The sum of rows with index values ‘A’, ‘B’, and ‘E’ for the assists column is 27.