How to Create a 3D Array in NumPy: Simple Guide

You can create a 3D array in NumPy by passing a list of lists of lists to np.array(). This creates an array with three dimensions representing depth, rows, and columns.

Syntax

To create a 3D array, use np.array() with nested lists. The outer list represents the depth (number of 2D arrays), each inner list represents rows, and the innermost lists represent columns.

• np.array(data): Converts nested lists into a NumPy array.
• data: A list of lists of lists for 3D structure.

python

import numpy as np

array_3d = np.array([
    [[1, 2, 3], [4, 5, 6]],
    [[7, 8, 9], [10, 11, 12]]
])

Example

This example creates a 3D array with 2 layers, each containing 2 rows and 3 columns. It then prints the array and its shape to show the dimensions.

python

import numpy as np

array_3d = np.array([
    [[1, 2, 3], [4, 5, 6]],
    [[7, 8, 9], [10, 11, 12]]
])

print("3D Array:")
print(array_3d)
print("Shape of array:", array_3d.shape)

Output

3D Array:
[[[ 1  2  3]
  [ 4  5  6]]

 [[ 7  8  9]
  [10 11 12]]]
Shape of array: (2, 2, 3)

Common Pitfalls

Common mistakes when creating 3D arrays include:

• Using uneven inner lists, which causes NumPy to create an array of objects instead of a numeric 3D array.
• Confusing the order of dimensions (depth, rows, columns).

Always ensure all inner lists have the same length to form a proper 3D array.

python

import numpy as np

# Wrong: uneven inner lists
wrong_array = np.array([
    [[1, 2], [3, 4, 5]],  # second inner list has 3 elements
    [[6, 7], [8, 9]]
])

print("Wrong array type:", wrong_array.dtype)

# Right: all inner lists have same length
right_array = np.array([
    [[1, 2, 3], [4, 5, 6]],
    [[7, 8, 9], [10, 11, 12]]
])

print("Right array type:", right_array.dtype)

Output

Wrong array type: object
Right array type: int64

Quick Reference

Concept	Description	Example
Create 3D array	Use np.array() with nested lists	np.array([[[1]]])
Shape	Returns (depth, rows, columns)	array.shape
Access element	Use three indices: depth, row, column	array[0, 1, 2]
Check type	Ensure numeric dtype, not object	array.dtype

Key Takeaways

Create a 3D array by passing nested lists to np.array().

Ensure all inner lists have the same length to avoid object arrays.

The shape of a 3D array is (depth, rows, columns).

Access elements using three indices: depth, row, and column.

Check the array's dtype to confirm it is numeric, not object.