How to flatten a 3D NumPy array and print its Strides?

Last update on December 21 2024 07:36:54 (UTC/GMT +8 hours)

Write a NumPy program that creates a 3D array of shape (2, 4, 4) and print the strides. Then, flatten it with ravel(order='F') and print the flattened array.

Sample Solution:

Python Code:

import numpy as np

# Create a 3D array of shape (2, 4, 4)
array_3d = np.array([[[ 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]]])

# Print the strides of the 3D array
print("Strides of the 3D array:", array_3d.strides)

# Flatten the array with ravel(order='F')
flattened_array = array_3d.ravel(order='F')

# Print the flattened array
print("Flattened array (column-major order):", flattened_array)

Output:

Strides of the 3D array: (64, 16, 4)
Flattened array (column-major order): [ 1 17  5 21  9 25 13 29  2 18  6 22 10 26 14 30  3 19  7 23 11 27 15 31
  4 20  8 24 12 28 16 32]

Explanation:

Import NumPy library: We start by importing the NumPy library to work with arrays.
Create a 3D array: We create a 3D array array_3d of shape (2, 4, 4) using np.array().
Print the strides: We print the strides of array_3d using the strides attribute. Strides indicate the number of bytes to step in each dimension when traversing an array.
Flatten with ravel(order='F'): We use the ravel() method with order='F' (column-major order) to flatten the 3D array into a 1D array, stored in flattened_array.
Print the flattened array: Finally, we print the flattened array flattened_array.

Python-Numpy Code Editor: