Langchain 的 Structured output parser

Langchain 的 Structured output parser

当您想要返回多个字段时,可以使用此输出解析器。虽然 Pydantic/JSON 解析器更强大,但我们最初尝试了仅具有文本字段的数据结构。

示例代码,

from langchain.output_parsers import StructuredOutputParser, ResponseSchema
from langchain.prompts import PromptTemplate, ChatPromptTemplate, HumanMessagePromptTemplate
from langchain.llms import OpenAI
from langchain.chat_models import ChatOpenAI

在这里我们定义了我们想要接收的响应模式。

示例代码,

response_schemas = [
    ResponseSchema(name="answer", description="answer to the user's question"),
    ResponseSchema(name="source", description="source used to answer the user's question, should be a website.")
]
output_parser = StructuredOutputParser.from_response_schemas(response_schemas)

现在,我们得到一个字符串,其中包含如何格式化响应的说明,然后将其插入提示符中。

format_instructions = output_parser.get_format_instructions()
prompt = PromptTemplate(
    template="answer the users question as best as possible.\n{format_instructions}\n{question}",
    input_variables=["question"],
    partial_variables={"format_instructions": format_instructions}
)

我们现在可以使用它来格式化提示以发送到语言模型,然后解析返回的结果。

model = OpenAI(temperature=0)
_input = prompt.format_prompt(question="what's the capital of france?")
output = model(_input.to_string())
output_parser.parse(output)

输出结果,

    {'answer': 'Paris',
     'source': 'https://www.worldatlas.com/articles/what-is-the-capital-of-france.html'}

这是在聊天模型中使用它的示例,

chat_model = ChatOpenAI(temperature=0)
prompt = ChatPromptTemplate(
    messages=[
        HumanMessagePromptTemplate.from_template("answer the users question as best as possible.\n{format_instructions}\n{question}")  
    ],
    input_variables=["question"],
    partial_variables={"format_instructions": format_instructions}
)
_input = prompt.format_prompt(question="what's the capital of france?")
output = chat_model(_input.to_messages())
output_parser.parse(output.content)

输出结果,

    {'answer': 'Paris', 'source': 'https://en.wikipedia.org/wiki/Paris'}

完结!

猜你喜欢

转载自blog.csdn.net/engchina/article/details/131873603