NumPy: Multiply an array of dimension by an array with dimensions
Multiply arrays of compatible dimensions.
Write a NumPy program to multiply an array of dimension (2,2,3) by an array with dimensions (2,2).
Sample Solution:
Python Code:
# Importing the NumPy library
import numpy as np
# Creating a NumPy array 'nums1' consisting of ones with shape (2, 2, 3)
nums1 = np.ones((2, 2, 3))
# Creating another NumPy array 'nums2' consisting of threes with shape (2, 2)
nums2 = 3 * np.ones((2, 2))
# Displaying the original array 'nums1'
print("Original array:")
print(nums1)
# Performing element-wise multiplication of 'nums1' and 'nums2' along the second axis using None to add a new axis
new_array = nums1 * nums2[:, :, None]
# Displaying the new array obtained after the multiplication
print("\nNew array:")
print(new_array)
Sample Output:
Original array: [[[1. 1. 1.] [1. 1. 1.]] [[1. 1. 1.] [1. 1. 1.]]] New array: [[[3. 3. 3.] [3. 3. 3.]] [[3. 3. 3.] [3. 3. 3.]]]
Explanation:
nums1 = np.ones((2,2,3)): Creates a 3D NumPy array nums1 of shape (2, 2, 3) filled with ones.
nums2 = 3*np.ones((2,2)): Creates a 2D NumPy array nums2 of shape (2, 2) filled with threes.
new_array = nums1 * nums2[:,:,None]
In the above code -
- nums2[:,:,None]: Adds a new axis to nums2 to make it a 3D array with shape (2, 2, 1). This is done to match the shape of nums1 for broadcasting.
- new_array = nums1 * nums2[:,:,None]: Performs element-wise multiplication of nums1 and the reshaped nums2, using broadcasting. Since nums1 contains ones and nums2 contains threes, the resulting new_array will be a 3D array of shape (2, 2, 3) filled with threes.
Python-Numpy Code Editor: