提升LangChain查询分析器：如何在提示中添加示例

nseejrukjhad

于 2024-10-03 17:32:37 发布

阅读量225

点赞数 3

文章标签： langchain windows python

本文链接：https://blog.csdn.net/nseejrukjhad/article/details/142693929

版权

引言

在开发使用大语言模型（LLM）的查询分析器时，我们常常面临一个挑战：模型可能无法准确理解复杂的查询。为了解决这一问题，我们可以在提示中添加示例，以指导模型更好地进行查询分析。本文将介绍如何为LangChain的YouTube视频查询分析器添加示例，以提高其性能。

主要内容

安装与环境设置

首先，我们需要安装必要的依赖项并设置环境变量。

# 安装依赖项
%pip install -qU langchain-core langchain-openai

# 设置环境变量
import getpass
import os

os.environ["OPENAI_API_KEY"] = getpass.getpass()

# 可选：启用LangSmith追踪
# os.environ["LANGCHAIN_TRACING_V2"] = "true"
# os.environ["LANGCHAIN_API_KEY"] = getpass.getpass()

定义查询架构

我们将定义一个查询架构，其中包含一个用于存储细分问题的sub_queries字段。

from typing import List, Optional
from langchain_core.pydantic_v1 import BaseModel, Field

sub_queries_description = """\
If the original question contains multiple distinct sub-questions..."""

class Search(BaseModel):
    query: str = Field(..., description="Primary similarity search query...")
    sub_queries: List[str] = Field(default_factory=list, description=sub_queries_description)
    publish_year: Optional[int] = Field(None, description="Year video was published")

查询生成

使用LangChain的工具来生成查询。

from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_core.runnables import RunnablePassthrough
from langchain_openai import ChatOpenAI

system = """You are an expert at converting user questions into database queries..."""

prompt = ChatPromptTemplate.from_messages([
    ("system", system),
    MessagesPlaceholder("examples", optional=True),
    ("human", "{question}"),
])

llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)
structured_llm = llm.with_structured_output(Search)
query_analyzer = {"question": RunnablePassthrough()} | prompt | structured_llm

添加示例优化提示

为了更细致地分解查询，我们需要在提示中添加具体示例。

examples = []

question = "What's chat langchain, is it a langchain template?"
query = Search(query="What is chat langchain and is it a langchain template?", sub_queries=["What is chat langchain", "What is a langchain template"])
examples.append({"input": question, "tool_calls": [query]})

#... (添加更多示例)

from langchain_core.messages import AIMessage, BaseMessage, HumanMessage, SystemMessage, ToolMessage

def tool_example_to_messages(example: Dict) -> List[BaseMessage]:
    messages: List[BaseMessage] = [HumanMessage(content=example["input"])]
    # 处理工具调用的消息转换
    return messages

example_msgs = [msg for ex in examples for msg in tool_example_to_messages(ex)]

将示例整合到查询分析器中

query_analyzer_with_examples = (
    {"question": RunnablePassthrough()}
    | prompt.partial(examples=example_msgs)
    | structured_llm
)

# 测试查询分析器
query_analyzer_with_examples.invoke(
    "what's the difference between web voyager and reflection agents? do both use langgraph?"
)