Stable Diffusion overview in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the main library used for Stable Diffusion.

Prompt Engineering / GenAI

import [1]

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

Atorch

Btensorflow

Cnumpy

Dpandas

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing tensorflow which is another deep learning library but not the main one used here.

Choosing numpy or pandas which are not for deep learning.

The torch library is used for Stable Diffusion because it supports deep learning models and GPU acceleration.

2fill in blank

medium

Complete the code to load a pre-trained Stable Diffusion model using the diffusers library.

Prompt Engineering / GenAI

from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained([1])

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

A'gpt2'

B'bert-base-uncased'

C'resnet50'

D'stable-diffusion-v1-4'

Attempts:

3 left

💡 Hint

Common Mistakes

Using model names for other AI models like GPT or BERT.

Using image classification model names like resnet50.

The 'stable-diffusion-v1-4' is a common pre-trained model name for Stable Diffusion in the diffusers library.

3fill in blank

hard

Fix the error in the code to generate an image from a text prompt using the pipeline.

Prompt Engineering / GenAI

image = pipeline([1]="a beautiful sunset over mountains").images[0]

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

Aprompt

Bdescription

Cinput

Dtext

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'text' or 'input' which are not valid parameter names here.

Using 'description' which is not recognized by the pipeline.

The correct argument name for the text input in the pipeline is prompt.

4fill in blank

hard

Fill both blanks to convert the generated image to a numpy array and display its shape.

Prompt Engineering / GenAI

import numpy as np
image_array = np.[1](image)
print(image_array.[2])

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

Aarray

Bshape

Csize

Dtolist

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'size' which gives total elements, not shape.

Using 'tolist' which converts array to list, not what is asked.

Use np.array to convert the image to a numpy array, and shape to get its dimensions.

5fill in blank

hard

Fill all three blanks to save the generated image to a file named 'output.png'.

Prompt Engineering / GenAI

from PIL import Image
img = Image.[1](image_array)
img.[2]('output.png', [3]='PNG')

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

Afromarray

Bsave

Cformat

Dopen

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'open' instead of 'fromarray' to create the image.

Not specifying the format or using wrong method to save.

Use Image.fromarray to create an image from the array, then save it with the format set to 'PNG'.