Strides and how data is accessed in NumPy - Mini Project: Build & Apply

Understanding Strides and How Data is Accessed in NumPy

📖 Scenario: Imagine you have a small photo represented as a grid of pixels. Each pixel has a brightness value. You want to understand how NumPy stores and accesses this pixel data efficiently using strides.

🎯 Goal: You will create a NumPy array representing pixel brightness, check its strides, and then access the data in different ways to see how strides affect data access.

📋 What You'll Learn

Create a 2D NumPy array with exact values

Check and store the strides of the array

Use slicing to create a sub-array and observe its strides

Print the strides and the sub-array to see how data is accessed

💡 Why This Matters

🌍 Real World

Understanding strides helps in optimizing image processing and scientific computing tasks where memory layout affects performance.

💼 Career

Data scientists and engineers often manipulate large arrays; knowing strides helps write efficient code and debug memory-related issues.

1

Create a 2D NumPy array

Import NumPy as np and create a 2D NumPy array called pixels with these exact values: [[10, 20, 30], [40, 50, 60], [70, 80, 90]].

NumPy

# Import numpy and create the pixels array
# Your code here

Need a hint?

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

2

Check the strides of the array

Create a variable called pixel_strides and set it to the strides of the pixels array using pixels.strides.

NumPy

import numpy as np
pixels = np.array([[10, 20, 30], [40, 50, 60], [70, 80, 90]])
# Get the strides of the pixels array
# Your code here

Need a hint?

Use the .strides attribute of the NumPy array.

3

Create a sub-array using slicing and check its strides

Create a new variable called sub_pixels by slicing the pixels array to get the first two rows and the last two columns. Then create a variable called sub_strides and set it to the strides of sub_pixels.

NumPy

import numpy as np
pixels = np.array([[10, 20, 30], [40, 50, 60], [70, 80, 90]])
pixel_strides = pixels.strides
# Slice pixels to get first two rows and last two columns
# Then get strides of the sub-array
# Your code here

Need a hint?

Use slicing pixels[:2, 1:3] to get the sub-array.

4

Print the strides and the sub-array

Print the variables pixel_strides, sub_strides, and sub_pixels each on its own line to see how the data is accessed.

NumPy

import numpy as np
pixels = np.array([[10, 20, 30], [40, 50, 60], [70, 80, 90]])
pixel_strides = pixels.strides
sub_pixels = pixels[:2, 1:3]
sub_strides = sub_pixels.strides
# Print pixel_strides, sub_strides, and sub_pixels
# Your code here

Need a hint?

Use three print() statements, one for each variable.