12/4/24, 11:18 AM Sha_LS_13a_and_13b.

ipynb - Colab

Generative AI CSE3348
Name: Pankaj Silot
20211cse0730
7CSE15

https://colab.research.google.com/drive/1ZvDUeFYe5dyDR219frV3OwvKw2YcswRC#scrollTo=fsqsjMjHNReB&printMode=true 1/5
12/4/24, 11:18 AM Sha_LS_13a_and_13b.ipynb - Colab

import os

# Set the OpenAI API key

os.environ['OPENAI_API_KEY'] = 'sk-VTZGcU2zYRjmcUOEUNqG4J9HUvwMlMJg_WGdU

api_key = os.getenv('OPENAI_API_KEY')

print(api_key) # Check if the key is successfully retrieved

sk-VTZGcU2zYRjmcUOEUNqG4J9HUvwMlMJg_WGdUa01WST3BlbkFJcC_CHLnJDt26s

!pip install openai

Requirement already satisfied: openai in /usr/local/lib/python3.10

Requirement already satisfied: anyio<5,>=3.5.0 in /usr/local/lib/p
Requirement already satisfied: distro<2,>=1.7.0 in /usr/local/lib/
Requirement already satisfied: httpx<1,>=0.23.0 in /usr/local/lib/
Requirement already satisfied: jiter<1,>=0.4.0 in /usr/local/lib/p
Requirement already satisfied: pydantic<3,>=1.9.0 in /usr/local/li
Requirement already satisfied: sniffio in /usr/local/lib/python3.1
Requirement already satisfied: tqdm>4 in /usr/local/lib/python3.10
Requirement already satisfied: typing-extensions<5,>=4.11 in /usr/
Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3
Requirement already satisfied: exceptiongroup in /usr/local/lib/py
Requirement already satisfied: certifi in /usr/local/lib/python3.1
Requirement already satisfied: httpcore==1.* in /usr/local/lib/pyt
Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/p
Requirement already satisfied: annotated-types>=0.6.0 in /usr/loca
Requirement already satisfied: pydantic-core==2.23.4 in /usr/local

import openai # Import an API client library

client = openai # Here, client is defined and assigned

client.api_key = "sk-VTZGcU2zYRjmcUOEUNqG4J9HUvwMlMJg_WGdUa01WST3Bl
bkFJc

import openai
import requests
from PIL import Image
import io

# Set up your OpenAI API key

openai.api_key = 'sk-VTZGcU2zYRjmcUOEUNqG4J9HUvwMlMJg_WGdUa01WST3Bl
bkF

def generate_image(prompt, n=1, size="1024x1024"):

"""
Generate an image using DALL-E 3

:param prompt: Text description of the image

:param n: Number of images to generate
:param size: Size of the image
:return: List of image URLs
"""
try:
response = client.images.generate
(
model="dall-e-3",
prompt=prompt,
n=n,
size=size
)

https://colab.research.google.com/drive/1ZvDUeFYe5dyDR219frV3OwvKw2YcswRC#scrollTo=fsqsjMjHNReB&printMode=true 2/5
12/4/24, 11:18 AM Sha_LS_13a_and_13b.ipynb - Colab

https://colab.research.google.com/drive/1ZvDUeFYe5dyDR219frV3OwvKw2YcswRC#scrollTo=fsqsjMjHNReB&printMode=true 3/5
12/4/24, 11:18 AM Sha_LS_13a_and_13b.ipynb - Colab
# Importing PyTorch library, for building and trai
ning neural networks
import torch

# Importing StableDiffusionPipeline to use pre-tra

ined Stable Diffusio
from diffusers import StableDiffusionPipeline

# Image is a class for the PIL module to visualizeimages in a Python

from PIL import Image

The cache for model files in Transformers v4.22.0 has been updated
0/0 [00:00<?, ?it/s]

# Creating pipeline
pipeline = StableDiffusionPipeline.from_pretrained
("CompVis/stable-dif
torch_dtype=torch.fl

/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_aut
The secret `HF_TOKEN` does not exist in your Colab secrets.
To authenticate with the Hugging Face Hub, create a token in your
You will be able to reuse this secret in all of your notebooks.
Please note that authentication is recommended but still optional
warnings.warn(
model_index.json: 100% 541/541 [00:00<00:00, 40.2kB/s]

Fetching 16 files: 100% 16/16 [00:37<00:00, 2.35s/it]

model.safetensors: 100% 492M/492M [00:11<00:00, 73.1MB/s]

text_encoder/config.json: 100% 592/592 [00:00<00:00, 6.83kB/s]

(…)ature_extractor/preprocessor_config.json: 100% 342/342 [00:00<00:00, 4.31kB/s

scheduler/scheduler_config.json: 100% 313/313 [00:00<00:00, 3.58kB/s]

safety_checker/config.json: 100% 4.56k/4.56k [00:00<00:00, 56.5kB/s]

tokenizer/merges.txt: 100% 525k/525k [00:00<00:00, 4.14MB/s]

model.safetensors: 100% 1.22G/1.22G [00:24<00:00, 16.2MB/s]

(…)kpoints/scheduler_config - 209/209 [00:00<00:00, 2.35kB/s]

checkpoint.json: 100%

diffusion_pytorch_model.safetensors: 100% 3.44G/3.44G [00:36<00:00, 223MB/s]

tokenizer/vocab.json: 100% 1.06M/1.06M [00:00<00:00, 5.21MB/s]

tokenizer/tokenizer_config.json: 100% 806/806 [00:00<00:00, 8.85kB/s]

tokenizer/special_tokens_map.json: 100% 472/472 [00:00<00:00, 4.75kB/s]

unet/config.json: 100% 743/743 [00:00<00:00, 7.79kB/s]

vae/config.json: 100% 551/551 [00:00<00:00, 13.8kB/s]

diffusion_pytorch_model.safetensors: 100% 335M/335M [00:06<00:00, 56.8MB/s]

# Defining function for the creation of a grid ofimages

def image_grid(imgs, rows, cols):
assert len(imgs) == rows*cols

w, h = imgs[0].size
grid = Image.new('RGB', size = (cols*w,
rows * ))
w
grid_w, grid_h = grid.size

for i, img in enumerate(imgs):

grid.paste(img, box = (i%cols*w, i // cols*h))
return grid

https://colab.research.google.com/drive/1ZvDUeFYe5dyDR219frV3OwvKw2YcswRC#scrollTo=fsqsjMjHNReB&printMode=true 4/5
12/4/24, 11:18 AM Sha_LS_13a_and_13b.ipynb - Colab

# Moving pipeline to GPU

# Check if CUDA is available before moving the pip
eline
if torch.cuda.is_available():
pipeline = pipeline.to('cuda')
else:
print("CUDA is not available. Using CPU instead."
)

n_images = 6 # Let's generate 6 images based on the prompt belo

w
prompt = ['Sunset on a beach'] * n_images

images = pipeline(prompt).images

grid = image_grid(images, rows=2, cols = 3)

grid

100% 50/50 [00:37<00:00, 1.37it/s]

Start coding or generate with AI.

https://colab.research.google.com/drive/1ZvDUeFYe5dyDR219frV3OwvKw2YcswRC#scrollTo=fsqsjMjHNReB&printMode=true 5/5