Creating and reshaping a 4D NumPy array

Last update on April 29 2025 13:01:50 (UTC/GMT +8 hours)

17. 4D Array Multi-Reshape

Write a NumPy program that creates a 4D array of shape (2, 2, 3, 3), reshape it into a 2D array, and then reshape it back to the original shape. Print all the intermediate arrays.

Sample Solution:

Python Code:

import numpy as np

# Step 1: Create a 4D array of shape (2, 2, 3, 3)
original_array = np.random.rand(2, 2, 3, 3)
print("Original 4D array:\n", original_array)

# Step 2: Reshape the 4D array into a 2D array
reshaped_2d_array = original_array.reshape(-1, 9)
print("\nReshaped 2D array:\n", reshaped_2d_array)

# Step 3: Reshape the 2D array back to the original 4D shape
reshaped_back_to_4d = reshaped_2d_array.reshape(2, 2, 3, 3)
print("\nReshaped back to 4D array:\n", reshaped_back_to_4d)

Output:

Original 4D array:
 [[[[0.87825329 0.92814202 0.72976418]
   [0.85342904 0.1590352  0.36378863]
   [0.6212089  0.2612585  0.70702319]]

  [[0.51610167 0.46228402 0.84994689]
   [0.70451938 0.01329465 0.86950388]
   [0.32117702 0.92568944 0.86945406]]]


 [[[0.90918915 0.41516567 0.77052305]
   [0.34892452 0.11246739 0.72093766]
   [0.22318908 0.00657031 0.06388555]]

  [[0.61764261 0.84885538 0.49016637]
   [0.46874106 0.90037212 0.34975796]
   [0.63524874 0.59394007 0.12072371]]]]

Reshaped 2D array:
 [[0.87825329 0.92814202 0.72976418 0.85342904 0.1590352  0.36378863
  0.6212089  0.2612585  0.70702319]
 [0.51610167 0.46228402 0.84994689 0.70451938 0.01329465 0.86950388
  0.32117702 0.92568944 0.86945406]
 [0.90918915 0.41516567 0.77052305 0.34892452 0.11246739 0.72093766
  0.22318908 0.00657031 0.06388555]
 [0.61764261 0.84885538 0.49016637 0.46874106 0.90037212 0.34975796
  0.63524874 0.59394007 0.12072371]]

Reshaped back to 4D array:
 [[[[0.87825329 0.92814202 0.72976418]
   [0.85342904 0.1590352  0.36378863]
   [0.6212089  0.2612585  0.70702319]]

  [[0.51610167 0.46228402 0.84994689]
   [0.70451938 0.01329465 0.86950388]
   [0.32117702 0.92568944 0.86945406]]]


 [[[0.90918915 0.41516567 0.77052305]
   [0.34892452 0.11246739 0.72093766]
   [0.22318908 0.00657031 0.06388555]]

  [[0.61764261 0.84885538 0.49016637]
   [0.46874106 0.90037212 0.34975796]
   [0.63524874 0.59394007 0.12072371]]]]

Explanation:

Create a 4D array: A 4D array of shape (2, 2, 3, 3) is created using np.random.rand().
Reshape to 2D: The 4D array is reshaped into a 2D array with shape (-1, 9), flattening the first three dimensions into a single dimension.
Reshape back to 4D: The 2D array is reshaped back to the original 4D shape (2, 2, 3, 3).

For more Practice: Solve these Related Problems:

Write a NumPy program to create a 4D array, reshape it into a 2D array, perform an arithmetic operation, and reshape it back to 4D.
Write a NumPy program to perform multiple sequential reshapes on a 4D array and print the shapes and strides at each step.
Write a NumPy program to verify data integrity by reshaping a 4D array into 2D and then reconstructing the original 4D shape.
Write a NumPy program to compare the strides of a 4D array before and after reshaping it to 2D and back.

Go to:

Previous: How to slice a Sub-array from a reshaped 1D NumPy array and print strides?
Next: Creating and reshaping a 2D NumPy array.

Python-Numpy Code Editor:

Have another way to solve this solution? Contribute your code (and comments) through Disqus.