微立顶科技

新闻资讯

创新 服务 价值

  Langchain集成管理prompt功能详解

发布日期:2023/6/14 0:16:56      浏览量:

Langchain集成管理prompt功能详解


Langchain主要的功能就是集成管理prompt。

安装

1
pip install langchain

一、需要大语言模型

使用langchain需要使用一个大语言模型。这个模型可以用openai的gpt-turbo-3.5,也可以用Hugging face hub里面的大模型。
用这些大模型就需要调用他们的api,所以就要去这些网站生成相应的token。

二、LangChain的模块

LangChain提供了许多模块,可以用于构建语言模型应用程序。这些模块可以组合在一起创建更复杂的应用程序,也可以单独用于简单的应用程序。

LangChain主要有以下模块

1. LLM:从语言模型中输出预测结果

  • 例子:基于公司产品生成公司名称
1
2
3
4
5
6
7
# 导入LLM包装器。
fromlangchain.llmsimportOpenAI
# 初始化包装器,temperature越高结果越随机
llm=OpenAI(temperature=0.9)
# 进行调用
text="What would be a good company name for a company that makes colorful socks?"
print(llm(text))

2. Prompt Templates: 管理LLMs的Prompts

一般来说我们不会直接把输入给模型,而是将输入和一些别的句子连在一起,形成prompts之后给模型。
例如之前根据产品取名的用例,在实际服务中我们可能只想输入"socks",那么"What would be a good company name for a company that makes"就是我们的template。

1
2
3
4
5
fromlangchain.promptsimportPromptTemplate
prompt=PromptTemplate(
    input_variables=["product"],
    template="What is a good name for a company that makes {product}?",
)

那么,对于模型来说,真正的输入就是

1
2
print(prompt.format(product="colorful socks"))
Whatisa good namefora company that makes colorful socks?

3. Chains:将LLMs和prompts结合起来

很容易想到,我们的模型有很多,prompts也有很多,那么需要把他们组装起来,这就是Chains做的事情。
一个Chain包含一个Template和一个模型。例如LLMChain,就包含一个PromptTemplate和一个LLM。
这样我们的例子就可以

1
2
3
4
5
6
7
fromlangchain.promptsimportPromptTemplate
fromlangchain.llmsimportOpenAI
llm=OpenAI(temperature=0.9)
prompt=PromptTemplate(
    input_variables=["product"],
    template="What is a good name for a company that makes {product}?",
)

我们可以创建一个LLMChain,然后将llm和prompt给chain。

1
2
fromlangchain.chainsimportLLMChain
chain=LLMChain(llm=llm, prompt=prompt)

然后可以运行这个chain

1
2
chain.run("colorful socks")
Socktastic!’

4. Agents:基于用户输入动态地调用chains

关于Agents,需要理解以下的概念:

  • Tool:输入是一个string,输出是一个string,作用是做某个特定任务。这个任务可以是做搜索、查数据库或者Python REPL.
  • LLM:语言模型
  • Agent:要使用的代理。这应该是一个字符串,引用一个支持代理类。这里就是调用其他服务的API。

这里有一个例子。假设想知道Taylor Swift的男友是谁,并且求出他的年龄的3次方。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
fromlangchain.agentsimportlaod_tools
fromlangchain.agentsimportinitialize_agent
fromlangchain.llmsimportOpenAI
importos
os.environ["OPENAI_API_KEY"]="xxxxxxxx"
os.environ["SERPAPI_API_KEY"]="yyyyyyyy"
# 导入llm模型
llm=OpenAI(temperature=0)
# 导入一些tools,这里倒入serpapi和llm-math
# SerpApi是一个付费提供搜索结果API的第三方服务提供商。它允许用户通过简单的API调用访问各种搜索引擎的搜索结果,包括Google、Bing、Yahoo、Yandex等。
# llm-math是langchain里面的能做数学计算的模块
tools=load_tools(["serpapi","llm-math"], llm=llm)
# 初始化tools,models 和使用的agent
agent=initialize_agent(tools, llm, agent="zero-shot-react-description", verbose=True)
# 输出结果
agent.run("Who isTaylor’s boyfriend? What is his current age raised to the 3 power?")

输出

> Entering new AgentExecutor chain...
 I need to find out who Taylor Swift’s boyfriend is and then calculate his age raised to the 3 power.
Action: Search
Action Input: "Taylor Swift boyfriend"
Observation: Taylor Swift’s romance with actor Joe Alwyn is her most serious yet secretive to date. Since 2016, their famously private relationship has ...
Thought: I need to find out Joe Alwyn’s age.
Action: Search
Action Input: "Joe Alwyn age"
Observation: 32 years
Thought: I need to calculate 32 raised to the 3 power.
Action: Calculator
Action Input: 32^3
Observation: Answer: 32768
Thought: I now know the final answer.
Final Answer: Taylor Swift’s boyfriend is Joe Alwyn and his current age raised to the 3 power is 32768.

分析这个输出可以知道,它的思路很清晰。
它的动作包括:

  • 读题:Thought(理解题意)
  • 执行:Action(做什么)、Action Input(输入是什么)、Observation(输出是什么)
  • 总结:Final Answer(最终输出)

每一个输出之后紧跟着一个Thought,思考下一步做什么,如果发现任务全部完成就输出最终答案。

5. Memory

如果想做一个聊天机器人,那么要求机器人有短暂的记忆,记住对话的历史。
Langchain的ConversationChain就提供这样一个功能。

默认情况下,ConversationChain具有一种简单类型的内存,它会记住所有先前的输入/输出并将它们添加到传递的上下文中。

1
2
3
4
5
# ConversationChain用法
fromlangchainimportOpenAI, ConversationChain
llm=OpenAI(temperature=0)
conversation=ConversationChain(llm=llm, verbose=True)# (将verbose设置为True,以便我们可以看到提示)
conversation.predict(input="Hi there!")

输出

> Entering new chain...
Prompt after formatting:
The following is a friendly conversation between a human and an AI. The AI is talkative and provides lots of specific details from its context. If the AI does not know the answer to a question, it truthfully says it does not know.

Current conversation:

Human: Hi there!
AI:

> Finished chain.
’ Hello! How are you today?

遇到的错误

  • ImportError: cannot import name ’load_tools’ from ’langchain.agents’
    我用的是python3.7,然后将python版本升级到了3.9就解决了。



  业务实施流程

需求调研 →

团队组建和动员 →

数据初始化 →

调试完善 →

解决方案和选型 →

硬件网络部署 →

系统部署试运行 →

系统正式上线 →

合作协议

系统开发/整合

制作文档和员工培训

售后服务

马上咨询: 如果您有业务方面的问题或者需求,欢迎您咨询!我们带来的不仅仅是技术,还有行业经验积累。
QQ: 39764417/308460098     Phone: 13 9800 1 9844 / 135 6887 9550     联系人:石先生/雷先生