OpenAI embeddings API in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the OpenAI library.

Prompt Engineering / GenAI

import [1]

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

Aopenai

Btensorflow

Cnumpy

Dpandas

Attempts:

3 left

💡 Hint

Common Mistakes

Importing unrelated libraries like tensorflow or numpy.

Using uppercase letters in the import statement.

The OpenAI Python library is imported using import openai.

2fill in blank

medium

Complete the code to set your OpenAI API key.

Prompt Engineering / GenAI

openai.api_key = [1]

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

Aos.getenv('OPENAI_API_KEY')

B12345

C'my_api_key'

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Hardcoding the API key as a string.

Using an invalid key format.

Using os.getenv('OPENAI_API_KEY') safely loads the API key from environment variables.

3fill in blank

hard

Fix the error in the code to create an embedding for the text.

Prompt Engineering / GenAI

response = openai.Embedding.create(input=[1], model='text-embedding-3-small')

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

A'Hello world'

BNone

C123

D['Hello world']

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a single string instead of a list.

Passing a number or None as input.

The input parameter must be a list of strings, so ['Hello world'] is correct.

4fill in blank

hard

Fill both blanks to extract the embedding vector from the response.

Prompt Engineering / GenAI

embedding_vector = response['data'][[1]]['[2]']

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

A0

B1

Cembedding

Dvector

Attempts:

3 left

💡 Hint

Common Mistakes

Using index 1 instead of 0.

Using the wrong key like 'vector'.

The embedding is stored in response['data'][0]['embedding'].

5fill in blank

hard

Fill all three blanks to create an embedding and print its length.

Prompt Engineering / GenAI

response = openai.Embedding.create(input=[1], model=[2])
embedding = response['data'][0][[3]]
print(len(embedding))

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

A['OpenAI is great']

B'text-embedding-3-large'

C'embedding'

D'data'

Attempts:

3 left

💡 Hint

Common Mistakes

Passing input as a string instead of a list.

Using an incorrect model name.

Accessing the wrong key in the response.

We pass a list of strings as input, specify the correct model name, and access the 'embedding' key to get the vector.