Why augmentation multiplies training data in Computer Vision - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to apply a horizontal flip to an image using torchvision transforms.

Computer Vision

transform = torchvision.transforms.Compose([torchvision.transforms.[1]()])

Drag options to blanks, or click blank then click option'

ANormalize

BRandomHorizontalFlip

CToTensor

DResize

Attempts:

3 left

💡 Hint

Common Mistakes

Using Resize or Normalize instead of a flip transform.

The RandomHorizontalFlip transform flips images horizontally, creating new variations and thus multiplying training data.

2fill in blank

medium

Complete the code to create augmented data by rotating images randomly within 30 degrees.

Computer Vision

transform = torchvision.transforms.Compose([torchvision.transforms.RandomRotation([1])])

Drag options to blanks, or click blank then click option'

A30

B15

C45

D60

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing 15 or 45 degrees which do not match the instruction.

RandomRotation with 30 degrees rotates images randomly between -30 and +30 degrees, increasing data variety.

3fill in blank

hard

Fix the error in the code to correctly apply color jitter augmentation.

Computer Vision

transform = torchvision.transforms.ColorJitter(brightness=[1])

Drag options to blanks, or click blank then click option'

A0.5

B'0.5'

CTrue

D[0.5]

Attempts:

3 left

💡 Hint

Common Mistakes

Using a string instead of a float for brightness.

The brightness parameter expects a float number, not a string or list. So 0.5 is correct.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps each image filename to its augmented version using a horizontal flip.

Computer Vision

augmented_images = {filename: transform([1]) for filename, [2] in images.items()}

Drag options to blanks, or click blank then click option'

Aimage

Bimg

Cdata

Dpicture

Attempts:

3 left

💡 Hint

Common Mistakes

Using different variable names for the blanks or incorrect variable names.

The dictionary comprehension uses img as the input to transform and img as the variable for image data in the loop.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that filters images with width greater than 100 and applies a rotation augmentation.

Computer Vision

augmented = {name: transform([1]) for name, [2] in dataset.items() if [3].width > 100}

Drag options to blanks, or click blank then click option'

Aimg

Bimage

Ddata

Attempts:

3 left

💡 Hint

Common Mistakes

Using inconsistent variable names between the loop and condition.

The comprehension uses img as input to transform, img as loop variable, and checks img.width for filtering.