Prompt Engineering 提示词工程的实践技巧（六）

2025-02-22

字数统计: 1.6k字 | 阅读时长≈ 7分

本篇博客主要总结了 DeepLearning AI 公开课程 Prompt Engineering 提示词工程 第七章节 Expanding 扩充任务的内容，也可以理解为该章节的笔记。

扩充任务，顾名思义，就是给出一段文本，可以是主题或说明等，让 LLM 生成一段更长的文本，例如一篇文章、一份邮件。本章节还将引入一个模型参数，temperature 温度。这个参数允许我们调节模型的多样性程度。

0. 环境准备

之前的章节中，我们都没有单独列出调用 openai 的环境准备，因为它们都很简单，并且如果是要调用不同的大模型，API各有不同。因此，为了专注于 prompt ，之前的几个章节都将 setup 的环节略过了。但是在这个章节，我们将引入一个新的参数，temperature，来增加模型输出的多元化程度，因此在这里特地列出，和其他章节作一个区分。

import openai
import os

from dotenv import load_dotenv, find_dotenv
_ = load_dotenv(find_dotenv()) # read local .env file

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"]

1. AI 自动回复邮件

这个示例，演示了如何提供消费者的原始评论消息，指定语气和态度，自定义回复的电子邮件。

如下所示，提供了评论的情绪和评论原文。实际上，我们可以将任务拆分，在 prompt 中要求 LLM 先识别评论的情绪，再根据情绪进行回复。在这里，假定情绪识别任务已经完成，我们直接要求 LLM 根据情绪和评论撰写回复。

# given the sentiment from the lesson on "inferring",
# and the original customer message, customize the email
sentiment = "negative"

# review for a blender 一个搅拌机的评论
review = f"""
So, they still had the 17 piece system on seasonal sale for around $49 in the month of November, about half off, but for some reason (call it price gouging) around the second week of December the prices all went up to about anywhere from between $70-$89 for the same system. And the 11 piece system went up around $10 or so in price also from the earlier sale price of $29. So it looks okay, but if you look at the base, the part where the blade locks into place doesn’t look as good as in previous editions from a few years ago, but I plan to be very gentle with it (example, I crush very hard items like beans, ice, rice, etc. in the blender first then pulverize them in the serving size I want in the blender then switch to the whipping blade for a finer flour, and use the cross cutting blade first when making smoothies, then use the flat blade if I need them finer/less pulpy). Special tip when making smoothies, finely cut and freeze the fruits and vegetables (if using spinach-lightly stew soften the spinach then freeze until ready for use-and if making sorbet, use a small to medium sized food processor) that you plan to use that way you can avoid adding so much ice if at all-when making your smoothie. After about a year, the motor was making a funny noise. I called customer service but the warranty expired already, so I had to buy another one. FYI: The overall quality has gone done in these types of products, so they are kind of counting on brand recognition and consumer loyalty to maintain sales. Got it in about two days.
"""

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`.
Customer review: ```{review}```
Review sentiment: {sentiment}
"""
response = get_completion(prompt)
print(response)

## 输出
Dear valued customer,

Thank you for taking the time to share your detailed feedback with us. We are truly sorry to hear about the issues you experienced with the pricing changes and the quality of the product. We apologize for any inconvenience this may have caused you.

If you have any further concerns or would like to discuss this matter further, please feel free to reach out to our customer service team for assistance. We are here to help address any issues you may have encountered.

Thank you again for your feedback and for being a loyal customer. We appreciate your support.

AI customer agent

在这段 prompt 中，我们先指明 LLM 当前扮演的角色，是一个销售部门的AI助理，任务是给有价值的客户回复邮件。邮件内容会用```标识，需要根据消费者的评论，生成一封感谢邮件，如果评论态度是负面的，则道歉并告知会改进消费者服务，如果是正向或者中立的，就表示感谢。另外，还要 LLM 使用评论中的具体细节，用简明、专业的语气撰写邮件。

2. 使用温度参数控制多元化程度

接下来，我们将介绍 temperature 参数的作用。这个参数能影响模型输出的多样性，即 temperature 的值能影响模型输出的随机性。但究竟什么是模型输出的多样性呢？直接解释定义或许有些抽象，我们通过一个示例来看。

假如，我们输入一个短语，“我最喜欢的食物是……”，模型会预测最有可能的下一个单词是披萨（当然如果是中国大模型应该会是其他的答案），其次可能的是寿司和墨西哥卷。当 temperature = 0时，模型的多样性（随机性）最低，它将输出最有可能的、概率最高的答案，也就是披萨。但是当 temperature 被调高，它就可能会选择一个概率不那么高的单词，比如饺子。当温度更高时，甚至可能选择玉米卷。

下图就很生动地展示了 temperature 这个参数的作用情况。

当我们需要构建一个可预测的、可靠的系统时，维持默认的设定，temperature = 0即可。但如果希望模型更有创造性，更天马行空一点，获得更多样的输出，就可以调高 temperature 参数。在之前的章节中，我们使用的都是默认设置，即 temperature = 0。你也可以尝试对它赋值，并重新运行一模一样的 prompt，看看输出发生了哪些变化。

Prompt Engineering 提示词工程笔记汇总

参考资料

ChatGPT Prompt Engineering for Developers - DeepLearning.AI

B站可以找到搬运翻译的中文版本👆：《吴恩达 x OpenAI Prompt课程》

本文作者： Berry
本文链接： https://wuxinberry.github.io/2025/02/22/Prompt-tips6/
版权声明： 本博客所有文章除特别声明外，均采用 MIT 许可协议。转载请注明出处！