How to Use nsmallest in pandas to Get Smallest Values

Use DataFrame.nsmallest(n, columns) to get the n rows with the smallest values in specified columns. It returns a new DataFrame sorted by those columns in ascending order.

Syntax

The nsmallest method has this syntax:

• n: Number of rows to return with smallest values.
• columns: Column name or list of columns to sort by.
• keep: Which duplicates to keep if there are ties. Options: 'first' (default), 'last', or 'all'.

python

DataFrame.nsmallest(n, columns, keep='first')

Example

This example shows how to get the 3 rows with the smallest values in the 'Age' column.

python

import pandas as pd

data = {'Name': ['Alice', 'Bob', 'Charlie', 'David', 'Eva'],
        'Age': [25, 30, 22, 35, 28],
        'Score': [88, 92, 85, 90, 95]}

df = pd.DataFrame(data)

# Get 3 rows with smallest Age
smallest_ages = df.nsmallest(3, 'Age')
print(smallest_ages)

Output

      Name  Age  Score
2  Charlie   22     85
0    Alice   25     88
4      Eva   28     95

Common Pitfalls

Common mistakes include:

• Passing a column name that does not exist causes an error.
• Using nsmallest on multiple columns requires a list of column names, not a single string.
• Confusing nsmallest with sort_values — nsmallest is faster for selecting top rows.

python

import pandas as pd

data = {'A': [3, 1, 2], 'B': [9, 8, 7]}
df = pd.DataFrame(data)

# Wrong: passing a string instead of list for multiple columns
# df.nsmallest(2, 'A,B')  # This will raise an error

# Right: pass list of columns
result = df.nsmallest(2, ['A', 'B'])
print(result)

Output

   A  B
1  1  8
2  2  7

Key Takeaways

Use nsmallest(n, columns) to quickly get rows with smallest values in specified columns.

Pass a single column name as a string or multiple columns as a list to the columns parameter.

The method returns a new DataFrame sorted ascending by those columns.

Use the keep parameter to control which duplicates to keep when values tie.

Avoid passing invalid column names or wrong types to prevent errors.