How to Use iloc in pandas for Data Selection

Use iloc in pandas to select rows and columns by their integer position. It accepts row and column indices in the format df.iloc[row_index, column_index] to extract data by position, not label.

Syntax

The basic syntax of iloc is df.iloc[row_indexer, column_indexer]. You can use integers, slices, or lists of integers to specify which rows and columns to select.

• row_indexer: integer, list, or slice for rows by position
• column_indexer: integer, list, or slice for columns by position

Both row and column indexers are zero-based and exclusive for slices on the stop index.

python

df.iloc[0, 1]  # Selects first row, second column

df.iloc[1:3, 0:2]  # Selects rows 2-3 and columns 1-2

df.iloc[[0, 2], [1, 3]]  # Selects specific rows and columns by position

Example

This example shows how to select specific rows and columns using iloc on a sample DataFrame.

python

import pandas as pd

# Create sample DataFrame
data = {'Name': ['Alice', 'Bob', 'Charlie', 'David'],
        'Age': [25, 30, 35, 40],
        'City': ['NY', 'LA', 'Chicago', 'Houston']}
df = pd.DataFrame(data)

# Select first row, second column (Age)
age_first = df.iloc[0, 1]

# Select rows 1 to 2 and columns 0 to 1 (Name and Age for Bob and Charlie)
subset = df.iloc[1:3, 0:2]

(age_first, subset)

Output

(25,      Name  Age
1     Bob   30
2  Charlie   35)

Common Pitfalls

Common mistakes when using iloc include:

• Using labels instead of integer positions (use loc for labels).
• Forgetting that slicing with iloc excludes the stop index.
• Mixing row and column indexers incorrectly (always use df.iloc[row_indexer, column_indexer]).

python

import pandas as pd

data = {'A': [10, 20, 30], 'B': [40, 50, 60]}
df = pd.DataFrame(data)

# Wrong: Using label with iloc (raises error)
# df.iloc['A', 0]

# Right: Use integer position
value = df.iloc[0, 0]

# Correct: Slicing excludes stop index
subset_correct = df.iloc[0:2, 0:2]  # selects rows 0 and 1, not 2

(value, subset_correct)

Output

(10,    A   B
0  10  40
1  20  50)

Quick Reference

Usage	Description	Example
Single element	Select one cell by row and column position	df.iloc[0, 1]
Row slice	Select multiple rows, all columns	df.iloc[1:4, :]
Column slice	Select all rows, multiple columns	df.iloc[:, 0:2]
List of positions	Select specific rows and columns	df.iloc[[0,2], [1,3]]
Negative indices	Select from the end	df.iloc[-1, -1]

Key Takeaways

Use iloc to select data by integer position, not by label.

Row and column indices in iloc start at zero and slices exclude the stop index.

Always use df.iloc[row_indexer, column_indexer] with integers, slices, or lists.

For label-based selection, use df.loc instead of iloc.

Be careful not to mix labels and positions to avoid errors.