How to Use Boolean Indexing in NumPy for Easy Data Selection

In NumPy, boolean indexing lets you select elements from an array by using a boolean array of the same shape, where True means keep the element and False means discard it. You create a boolean condition on the array, then use it inside square brackets to get the filtered result.

Syntax

Boolean indexing uses a boolean array or condition inside square brackets to select elements from a NumPy array.

Syntax parts:

• array: Your original NumPy array.
• condition: A boolean array or expression that returns True or False for each element.
• array[condition]: Returns a new array with elements where the condition is True.

python

filtered_array = array[condition]

Example

This example shows how to select all numbers greater than 5 from a NumPy array using boolean indexing.

python

import numpy as np

array = np.array([1, 3, 7, 9, 2, 6])
condition = array > 5
filtered_array = array[condition]
print(filtered_array)

Output

[7 9 6]

Common Pitfalls

Common mistakes when using boolean indexing include:

• Using a boolean array of a different shape than the original array, which causes an error.
• Forgetting that the condition must be a boolean array, not just a number or string.
• Trying to assign values to a filtered array without using the original array, which does not change the original data.

python

import numpy as np

array = np.array([1, 2, 3, 4])

# Wrong: boolean array shape mismatch
try:
    wrong_condition = np.array([True, False])
    print(array[wrong_condition])
except Exception as e:
    print(f"Error: {e}")

# Right: boolean array matches shape
right_condition = array % 2 == 0
print(array[right_condition])

Output

Error: boolean index did not match indexed array along dimension 0; dimension is 4 but corresponding boolean dimension is 2
[2 4]

Quick Reference

Operation	Example	Description
Select elements > 5	array[array > 5]	Returns elements greater than 5
Select even numbers	array[array % 2 == 0]	Returns even elements
Select elements in range	array[(array > 2) & (array < 8)]	Returns elements between 2 and 8
Assign to filtered elements	array[array < 3] = 0	Sets elements less than 3 to zero

Key Takeaways

Boolean indexing uses a boolean array to select elements from a NumPy array.

The boolean array must have the same shape as the original array.

You can combine conditions with & (and), | (or) for complex filtering.

Boolean indexing returns a new array with only the selected elements.

Assigning to boolean indexed elements changes the original array.