如何为查询分析添加示例以提升精度_systemmessage csdn-优快云博客

在复杂的查询分析中，语言模型可能难以理解其在某些场景中的具体响应方式。为改进此类场景下的性能，我们可以在提示中添加示例来引导LLM（语言模型）。接下来，我将展示如何为我们在快速入门中构建的LangChain YouTube视频查询分析器添加示例。

技术背景介绍

为了创建一个能够处理复杂查询的智能分析器，我们通常需要引入一种机制，让模型理解如何根据上下文或问题格式提供响应。通过在提示中加入实例以及与之对应的理想输出，模型可以更好地解构和回答用户的问题。

核心原理解析

引入实例的核心原理是通过提供明确的输入与输出配对示例，指导模型了解如何处理类似的未见过的问题。实例作为指引性的模板，帮助模型在遇到相似问题时，能更好地拆解和组合生成答案。

代码实现演示

以下步骤说明如何为LangChain视频查询分析器添加示例。

设置环境

# 安装必要的库
%pip install -qU langchain-core langchain-openai

配置API密钥

我们将使用OpenAI服务：

import getpass
import os

os.environ["OPENAI_API_KEY"] = getpass.getpass()  # 输入您的API密钥

定义查询模式

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

class Search(BaseModel):
    query: str = Field(
        ...,
        description="Primary similarity search query applied to video transcripts.",
    )
    sub_queries: List[str] = Field(
        default_factory=list,
        description="包含原始问题的所有子问题的列表。"
    )
    publish_year: Optional[int] = Field(None, description="Year video was published")

查询生成

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. \
You have access to a database of tutorial videos about a software library for building LLM-powered applications. \
Given a question, return a list of database queries optimized to retrieve the most relevant results.

If there are acronyms or words you are not familiar with, do not try to rephrase them."""

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

query_analyzer.invoke(
    "what's the difference between web voyager and reflection agents? do both use langgraph?"
)

添加示例和优化提示

通过添加实例，我们可以进一步细化查询。以下是如何添加示例：

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

from langchain_core.messages import (
    AIMessage,
    BaseMessage,
    HumanMessage,
    SystemMessage,
    ToolMessage,
)
import uuid

def tool_example_to_messages(example):
    messages = [HumanMessage(content=example["input"])]
    openai_tool_calls = [
        {
            "id": str(uuid.uuid4()),
            "type": "function",
            "function": {
                "name": tool_call.__class__.__name__,
                "arguments": tool_call.json(),
            },
        }
        for tool_call in example["tool_calls"]
    ]
    messages.append(AIMessage(content="", additional_kwargs={"tool_calls": openai_tool_calls}))
    return messages

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

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

query_analyzer_with_examples.invoke(
    "what's the difference between web voyager and reflection agents? do both use langgraph?"
)