np.expand_dims() and np.squeeze() in NumPy - Mini Project: Build & Apply

Using np.expand_dims() and np.squeeze() in NumPy

📖 Scenario: Imagine you are working with image data in a simple computer vision project. Images are stored as arrays, but sometimes you need to add or remove extra dimensions to prepare the data for a model.

🎯 Goal: You will learn how to add a new dimension to a NumPy array using np.expand_dims() and how to remove single-dimensional entries using np.squeeze(). This helps in reshaping data correctly for processing.

📋 What You'll Learn

Create a 2D NumPy array called image with exact values

Create a variable called axis_to_expand to specify where to add a new dimension

Use np.expand_dims() to add a new dimension to image and save it as expanded_image

Use np.squeeze() to remove single-dimensional entries from expanded_image and save it as squeezed_image

Print the shapes of image, expanded_image, and squeezed_image to see the changes

💡 Why This Matters

🌍 Real World

In real-world data science, reshaping arrays is common when preparing data for machine learning models, especially in image processing and deep learning.

💼 Career

Understanding how to manipulate array dimensions is essential for data scientists and machine learning engineers to correctly format data inputs and outputs.

1

Create the initial 2D image array

Create a 2D NumPy array called image with these exact values: [[10, 20, 30], [40, 50, 60]]

NumPy

import numpy as np
# Create the 2D array called image with the exact values
# Your code here

Need a hint?

Use np.array() to create the array with the given nested list.

2

Set the axis to expand

Create a variable called axis_to_expand and set it to 0 to add a new dimension as the first axis

NumPy

import numpy as np
image = np.array([[10, 20, 30], [40, 50, 60]])
# Create axis_to_expand and set it to 0
# Your code here

Need a hint?

Just assign the number 0 to the variable axis_to_expand.

3

Add a new dimension using np.expand_dims()

Use np.expand_dims() with image and axis_to_expand to create expanded_image that has a new dimension added

NumPy

import numpy as np
image = np.array([[10, 20, 30], [40, 50, 60]])
axis_to_expand = 0
# Use np.expand_dims to add a new dimension and save as expanded_image
# Your code here

Need a hint?

Call np.expand_dims() with image and axis=axis_to_expand.

4

Remove single-dimensional entries using np.squeeze() and print shapes

Use np.squeeze() on expanded_image to create squeezed_image. Then print the shapes of image, expanded_image, and squeezed_image each on a new line.

NumPy

import numpy as np
image = np.array([[10, 20, 30], [40, 50, 60]])
axis_to_expand = 0
expanded_image = np.expand_dims(image, axis=axis_to_expand)
# Use np.squeeze on expanded_image to create squeezed_image
# Print the shapes of image, expanded_image, and squeezed_image
# Your code here

Need a hint?

Use np.squeeze() on expanded_image. Then print the .shape attribute of each array.