[LLM] ChatGPT Prompt Engineering for Developers - Expanding

 

ChatGPT Prompt Engineering for Developers

 

ChatGPT Prompt Engineering for Developers 

 

6 - Expanding  정리

 

Setup

import openai
import os

from dotenv import load_dotenv, find_dotenv
_ = load_dotenv(find_dotenv())

openai.api_key  = os.getenv('OPENAI_API_KEY')

def get_completion(prompt, model="gpt-3.5-turbo",temperature=0): # Andrew mentioned that the prompt/ completion paradigm is preferable for this class
    messages = [{"role": "user", "content": prompt}]
    response = openai.ChatCompletion.create(
        model=model,
        messages=messages,
        temperature=temperature, # this is the degree of randomness of the model's output
    )
    return response.choices[0].message["content"]

 

Expanding 

1) 고객 리뷰 자동응답  

주어진 Sentiment(감정 positive/negative)을 활용하여 고객 리뷰 작성

 

<prompt task>

You are a customer service AI assistant.

Your task is to send an email reply to a valued customer.

Given the customer email delimited by ```,  Generate a reply to thank the customer for their review.

If the sentiment is positive or neutral, thank them for  their review.

If the sentiment is negative, apologize and suggest that they can reach out to customer service.

Make sure to use specific details from the review.

Write in a concise and professional tone.

Sign the email as `AI customer agent`.

translate in korean.

 

Customer review: ```{review}```

Review sentiment: {sentiment}

 

 

 

+ Model의 Exploration(자유도) Expanding  / Temperature Randomness

Temperature를 높여 특정 키워드에 대한 언급 빈도를 랜덤성으로 추가한다. 

 

prompt = f"""
You are a customer service AI assistant.
Your task is to send an email reply to a valued customer.
Given the customer email delimited by ```, \
Generate a reply to thank the customer for their review.
If the sentiment is positive or neutral, thank them for \
their review.
If the sentiment is negative, apologize and suggest that \
they can reach out to customer service. 
Make sure to use specific details from the review.
Write in a concise and professional tone.
Sign the email as `AI customer agent`.
translate in korean.
Customer review: ```{review}```
Review sentiment: {sentiment}
"""
response = get_completion(prompt, temperature=0.7)
print(response)

 

 

temperature = 0.7

temperature = 1.0

 

Temperature의 정도에 따라 같은 parameter라도 다른 값이 나오는 것을 확인 할 수 있다.

 

temperature = 1.0