【langchain学习】使用Langchain生成多视角查询

nidieolc2024-08-04 22:22

使用Langchain生成多视角查询

导入所需库：

python 复制代码

from langchain.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser
from langchain_core.runnables import RunnablePassthrough
from config import llm

设置提示模板：

python 复制代码

prompt = ChatPromptTemplate.from_template(
    """
    You are an intelligent assistant. Your task is to generate 5 questions based on the provided question in different wording and different perspectives to retrieve relevant documents from a vector database. By generating multiple perspectives on the user question, your goal is to help the user overcome some of the limitations of the distance-based similarity search. Provide these alternative questions separated by newlines. Original question: {question}
    """
)

定义查询生成过程：

python 复制代码

generate_queries = (
    {"question": RunnablePassthrough()}
    | prompt
    | llm
    | StrOutputParser()
    | (lambda x: x.split("\n"))
)

调用生成多视角查询：

python 复制代码

result = generate_queries.invoke("温格高是如何赢下2023年环法自行车赛的？")
print(result)

示例输出：

python 复制代码

[
    "温格高在2023年环法自行车赛中采用了哪些策略？",
    "2023年环法自行车赛温格高获胜的关键因素是什么？",
    "温格高在2023年环法自行车赛中面对哪些挑战？",
    "温格高是如何准备2023年环法自行车赛的？",
    "2023年环法自行车赛温格高的表现有哪些亮点？"
]