How to Use nlargest in pandas to Get Top Values

Use DataFrame.nlargest(n, columns) to get the top n rows ordered by the specified columns. It returns a new DataFrame with the largest values sorted descending by default.

Syntax

The nlargest method selects the top n rows ordered by one or more columns.

• n: Number of rows to return.
• columns: Column name or list of column names to sort by.
• keep: Which duplicates to keep ('first', 'last', or 'all'). Default is 'first'.

python

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

Example

This example shows how to get the top 3 rows with the highest values in the 'score' column.

python

import pandas as pd

data = {'name': ['Alice', 'Bob', 'Charlie', 'David', 'Eva'],
        'score': [85, 92, 88, 91, 87]}
df = pd.DataFrame(data)

# Get top 3 rows by 'score'
top_scores = df.nlargest(3, 'score')
print(top_scores)

Output

    name  score
1    Bob     92
3  David     91
2 Charlie     88

Common Pitfalls

Common mistakes include:

• Using nlargest on columns with non-numeric data will cause errors.
• Passing multiple columns requires a list, not a string.
• Confusing nlargest with sort_values which sorts all rows.

python

import pandas as pd

data = {'name': ['Alice', 'Bob', 'Charlie'], 'score': [85, 92, 88]}
df = pd.DataFrame(data)

# Wrong: passing string instead of list for multiple columns
# df.nlargest(2, 'name')  # Works but not meaningful for strings

# Correct: use list for multiple columns
# df.nlargest(2, ['score'])  # Works fine

# Wrong: using nlargest on non-numeric column
# df.nlargest(2, 'name')  # Raises TypeError

Quick Reference

Summary tips for using nlargest:

• Use for numeric columns only.
• Specify n as the number of top rows you want.
• Pass column name as string or list of strings.
• Use keep='all' to keep all duplicates.

Key Takeaways

Use DataFrame.nlargest(n, columns) to get top n rows by column values.

Only numeric columns work correctly with nlargest.

Pass column names as string or list for multiple columns.

Use keep parameter to control duplicate handling.

nlargest returns rows sorted descending by the specified columns.