Image-to-image transformation in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load an image and convert it to a tensor for image-to-image transformation.

Prompt Engineering / GenAI

from PIL import Image
import torchvision.transforms as transforms

image = Image.open('input.jpg')
transform = transforms.Compose([
    transforms.ToTensor(),
])
tensor_image = [1](image)

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

AImage.open

Btransform

Ctransforms

DImage

Attempts:

3 left

💡 Hint

Common Mistakes

Using Image.open again instead of the transform.

Trying to call transforms directly without Compose.

The transform object applies the ToTensor() operation to convert the PIL image into a tensor.

2fill in blank

medium

Complete the code to define a simple convolutional layer for image-to-image transformation.

Prompt Engineering / GenAI

import torch.nn as nn

conv_layer = nn.Conv2d(in_channels=3, out_channels=16, kernel_size=[1], padding=1)

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

A3

B1

C5

D7

Attempts:

3 left

💡 Hint

Common Mistakes

Using kernel size 1 which is too small to capture features.

Using kernel size 7 which is large and uncommon here.

A kernel size of 3 is common for convolutional layers to capture local features effectively.

3fill in blank

hard

Fix the error in the code to apply a ReLU activation after the convolutional layer.

Prompt Engineering / GenAI

import torch.nn.functional as F

output = F.[1](conv_layer(input_tensor))

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

Arelu

Bsigmoid

Csoftmax

Dtanh

Attempts:

3 left

💡 Hint

Common Mistakes

Using sigmoid which is less common for convolutional activations.

Using softmax which is for classification outputs.

ReLU is the correct activation function to apply after convolution for non-linearity in image tasks.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps image filenames to their transformed tensors if the image size is greater than 256.

Prompt Engineering / GenAI

image_tensors = {filename: [1] for filename, image in images.items() if image.size[0] [2] 256}

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

Atransform(image)

Btransform

C>

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Using transform without calling it as a function.

Using '<' instead of '>' in the condition.

We apply the transform to the image and filter images with width greater than 256 pixels.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that maps image names to their grayscale tensors if the height is less than 128.

Prompt Engineering / GenAI

grayscale_images = {name: [1](image.convert([2])) for name, image in dataset.items() if image.size[1] [3] 128}

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

Atransform

B'L'

C<

D'RGB'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'RGB' instead of 'L' for grayscale conversion.

Using '>' instead of '<' in the size condition.

We convert images to grayscale with 'L', apply the transform, and filter images with height less than 128.