NumPy: Get all 2D diagonals of a 3D numpy array

Last update on August 29 2025 12:48:41 (UTC/GMT +8 hours)

Extract all 2D diagonals of a 3D array.

Write a NumPy program to get all 2D diagonals of a 3D numpy array.

Sample Solution:

Python Code:

# Importing necessary libraries
import numpy as np

# Creating a 3D NumPy array with dimensions 3x4x5 using arange and reshape methods
np_array = np.arange(3 * 4 * 5).reshape(3, 4, 5)

# Printing the original 3D NumPy array and its type
print("Original NumPy array:")
print(np_array)
print("Type: ", type(np_array))

# Extracting 2D diagonals from the 3D array with the specified axes using np.diagonal
result = np.diagonal(np_array, axis1=1, axis2=2)

# Printing the 2D diagonals and their type
print("\n2D diagonals: ")
print(result)
print("Type: ", type(result)) 

Sample Output:

Original Numpy array:
[[[ 0  1  2  3  4]
  [ 5  6  7  8  9]
  [10 11 12 13 14]
  [15 16 17 18 19]]

 [[20 21 22 23 24]
  [25 26 27 28 29]
  [30 31 32 33 34]
  [35 36 37 38 39]]

 [[40 41 42 43 44]
  [45 46 47 48 49]
  [50 51 52 53 54]
  [55 56 57 58 59]]]
Type:  <class 'numpy.ndarray'>

2D diagonals: 
[[ 0  6 12 18]
 [20 26 32 38]
 [40 46 52 58]]
Type:  <class 'numpy.ndarray'>

Explanation:

np_array = np.arange(3*4*5).reshape(3,4,5)

In the above code -

• np.arange(3*4*5) generates a 1D array of integers from 0 to (345)-1, which is from 0 to 59.
• reshape(3, 4, 5) reshapes the 1D array into a 3D array with dimensions 3x4x5.
• rp_array stores the created 3D array.

result = np.diagonal(np_array, axis1=1, axis2=2): This code computes the diagonal elements of the 3D array along the specified axes (axis1 and axis2). In this case, the diagonals are taken along the 2nd (axis1=1) and 3rd (axis2=2) dimensions of the array. For each element along the first dimension, the function picks the diagonal elements from the 4x5 sub-arrays. ‘result’ variable stores the computed diagonal elements.

Pictorial Presentation:

For more Practice: Solve these Related Problems:

• Write a NumPy program to extract the main 2D diagonal from each 2D slice of a 3D array using np.diagonal.
• Create a function that iterates over the first axis of a 3D array and returns an array of all 2D diagonals.
• Implement a solution that uses list comprehension to gather diagonals from each 2D sub-array and stacks them into a 2D array.
• Test the diagonal extraction on a 3D array with non-uniform sub-array sizes to ensure robustness.

Go to:

• NumPy Array Exercises Home ↩
• NumPy Exercises Home ↩

PREV : Convert Pandas DataFrame to NumPy array with headers.
NEXT : Count occurrences of sequences in a 2D array.

Python-Numpy Code Editor: