How to Drop Row in pandas: Simple Guide with Examples

To drop a row in pandas, use the DataFrame.drop() method with the row's index label or labels. You can also drop rows based on a condition by filtering the DataFrame.

Syntax

The basic syntax to drop rows by index is:

• DataFrame.drop(labels, axis=0, inplace=False)

Where:

• labels is the index label or list of labels of rows to drop.
• axis=0 means rows (default).
• inplace=False returns a new DataFrame without the rows; True modifies the original DataFrame.

python

df.drop(labels, axis=0, inplace=False)

Example

This example shows how to drop a single row by index label and multiple rows by a list of labels.

python

import pandas as pd

data = {'Name': ['Alice', 'Bob', 'Charlie', 'David'], 'Age': [25, 30, 35, 40]}
df = pd.DataFrame(data, index=['a', 'b', 'c', 'd'])

# Drop single row with index 'b'
df_single_drop = df.drop('b')

# Drop multiple rows with indexes 'a' and 'd'
df_multiple_drop = df.drop(['a', 'd'])

print('Original DataFrame:\n', df)
print('\nAfter dropping row with index b:\n', df_single_drop)
print('\nAfter dropping rows with indexes a and d:\n', df_multiple_drop)

Output

Original DataFrame:
    Name  Age
a  Alice   25
b    Bob   30
c Charlie   35
d  David   40

After dropping row with index b:
    Name  Age
a  Alice   25
c Charlie   35
d  David   40

After dropping rows with indexes a and d:
    Name  Age
b    Bob   30
c Charlie   35

Common Pitfalls

Common mistakes when dropping rows include:

• Forgetting that drop() returns a new DataFrame by default and does not change the original unless inplace=True is set.
• Using labels that do not exist in the index, which raises a KeyError.
• Confusing axis=0 (rows) with axis=1 (columns).

python

import pandas as pd

data = {'Name': ['Alice', 'Bob'], 'Age': [25, 30]}
df = pd.DataFrame(data, index=['a', 'b'])

# Wrong: trying to drop a non-existing index 'c' (raises KeyError)
# df.drop('c')

# Correct: use errors='ignore' to avoid error
new_df = df.drop('c', errors='ignore')

# Wrong: forgetting inplace=True means original df is unchanged
# df.drop('a')

# Correct: modify original df
df.drop('a', inplace=True)

print(new_df)
print(df)

Output

  Name  Age
a Alice   25
b   Bob   30

  Name  Age
b   Bob   30

Quick Reference

Method	Description	Example
drop(labels, axis=0, inplace=False)	Drop rows by index labels	df.drop('a')
drop(labels, axis=0, inplace=True)	Drop rows and modify original DataFrame	df.drop(['a','b'], inplace=True)
drop(labels, errors='ignore')	Avoid error if label not found	df.drop('x', errors='ignore')
Boolean filtering	Drop rows by condition	df = df[df['Age'] > 30]

Key Takeaways

Use df.drop() with row index labels to remove rows from a DataFrame.

Set inplace=True to modify the original DataFrame directly.

Use errors='ignore' to avoid errors when labels do not exist.

You can also drop rows by filtering with conditions instead of drop().

Remember axis=0 means rows; axis=1 means columns.