`langchain.chains.structured_output.base`.create_structured_output_runnable¶

langchain.chains.structured_output.base.create_structured_output_runnable(output_schema: Union[Dict[str, Any], Type[BaseModel]], llm: Runnable, prompt: Optional[BasePromptTemplate] = None, *, output_parser: Optional[Union[BaseOutputParser, BaseGenerationOutputParser]] = None, enforce_function_usage: bool = True, return_single: bool = True, mode: Literal['openai-functions', 'openai-tools', 'openai-json'] = 'openai-functions', **kwargs: Any) → Runnable[source]¶

[Deprecated] 创建一个可运行的程序，用于提取结构化输出。

参数：

output_schema：可以是字典或pydantic.BaseModel类。如果传入的是字典，: 则假定它已经是一个有效的JsonSchema。为了获得最佳结果，pydantic.BaseModels应该有描述模式代表什么以及参数描述的文档字符串。
llm：要使用的语言模型。假定支持OpenAI函数调用API: 如果模式是’openai-function’。假定支持OpenAI response_format 参数如果模式是’openai-json’。
prompt：传递给模型的BasePromptTemplate。如果模式是’openai-json’且: prompt具有输入变量’output_schema’，则给定的output_schema 将被转换为JsonSchema并插入到prompt中。
output_parser：用于解析模型输出的输出解析器。默认情况下: 将从函数类型中推断出来。如果传入pydantic.BaseModel，那么OutputParser将尝试使用pydantic类解析输出。否则模型输出将被解析为JSON。
mode：从模型中提取结构化输出的方式。如果是’openai-functions’: 则使用已弃用的’functions’，’function_call’模式进行OpenAI函数调用。如果是’openai-tools’，则使用最新的’tools’，’tool_choice’模式进行OpenAI函数调用。这比’openai-functions’更推荐。如果是’openai-json’，则使用OpenAI模型并将response_format设置为JSON。
enforce_function_usage：仅适用于模式为’openai-tools’或: ‘openai-functions’时。如果为True，则模型将被强制使用给定的输出模式。如果为False，则模型可以选择是否使用输出模式。
return_single：仅适用于模式为’openai-tools’时。是否返回一个结构化输出列表: 还是单个输出。如果为True且模型不返回任何结构化输出，则链输出为None。如果为False且模型不返回任何结构化输出则链输出为一个空列表。

**kwargs：额外的命名参数。

返回：

一个可运行的序列，将返回与给定output_schema匹配的结构化输出。

使用Pydantic模式的OpenAI工具示例（mode=’openai-tools’）：

from typing import Optional

from langchain.chains import create_structured_output_runnable
from langchain_openai import ChatOpenAI
from langchain_core.pydantic_v1 import BaseModel, Field


class RecordDog(BaseModel):
    '''记录有关狗的一些标识信息。'''

    name: str = Field(..., description="狗的名字")
    color: str = Field(..., description="狗的颜色")
    fav_food: Optional[str] = Field(None, description="狗喜欢的食物")

llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)
prompt = ChatPromptTemplate.from_messages(
    [
        ("system", "您是一个提取算法。请提取每个可能的实例"),
        ('human', '{input}')
    ]
)
structured_llm = create_structured_output_runnable(
    RecordDog,
    llm,
    mode="openai-tools",
    enforce_function_usage=True,
    return_single=True
)
structured_llm.invoke({"input": "Harry是一只胖胖的棕色比格犬，喜欢吃鸡肉"})
# -> RecordDog(name="Harry", color="brown", fav_food="chicken")

使用字典模式的OpenAI工具示例（mode=”openai-tools”）：

from typing import Optional

from langchain.chains import create_structured_output_runnable
from langchain_openai import ChatOpenAI


dog_schema = {
    "type": "function",
    "function": {
        "name": "record_dog",
        "description": "记录有关狗的一些标识信息。",
        "parameters": {
            "type": "object",
            "properties": {
                "name": {
                    "description": "狗的名字",
                    "type": "string"
                },
                "color": {
                    "description": "狗的颜色",
                    "type": "string"
                },
                "fav_food": {
                    "description": "狗喜欢的食物",
                    "type": "string"
                }
            },
            "required": ["name", "color"]
        }
    }
}


llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)
structured_llm = create_structured_output_runnable(
    dog_schema,
    llm,
    mode="openai-tools",
    enforce_function_usage=True,
    return_single=True
)
structured_llm.invoke("Harry是一只胖胖的棕色比格犬，喜欢吃鸡肉")
# -> {'name': 'Harry', 'color': 'brown', 'fav_food': 'chicken'}

OpenAI函数示例（mode=”openai-functions”）：

from typing import Optional

from langchain.chains import create_structured_output_runnable
from langchain_openai import ChatOpenAI
from langchain_core.pydantic_v1 import BaseModel, Field

class Dog(BaseModel):
    '''有关狗的标识信息。'''

    name: str = Field(..., description="狗的名字")
    color: str = Field(..., description="狗的颜色")
    fav_food: Optional[str] = Field(None, description="狗喜欢的食物")

llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)
structured_llm = create_structured_output_runnable(Dog, llm, mode="openai-functions")
structured_llm.invoke("Harry是一只胖胖的棕色比格犬，喜欢吃鸡肉")
# -> Dog(name="Harry", color="brown", fav_food="chicken")

带有提示的OpenAI函数示例：

from typing import Optional

from langchain.chains import create_structured_output_runnable
from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.pydantic_v1 import BaseModel, Field

class Dog(BaseModel):
    '''有关狗的标识信息。'''

    name: str = Field(..., description="狗的名字")
    color: str = Field(..., description="狗的颜色")
    fav_food: Optional[str] = Field(None, description="狗喜欢的食物")

llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)
structured_llm = create_structured_output_runnable(Dog, llm, mode="openai-functions")
system = '''从用户输入中提取提到的任何狗的信息。'''
prompt = ChatPromptTemplate.from_messages(
    [("system", system), ("human", "{input}"),]
)
chain = prompt | structured_llm
chain.invoke({"input": "Harry是一只胖胖的棕色比格犬，喜欢吃鸡肉"})
# -> Dog(name="Harry", color="brown", fav_food="chicken")

OpenAI JSON响应格式示例（mode=”openai-json”）：

from typing import Optional

from langchain.chains import create_structured_output_runnable
from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.pydantic_v1 import BaseModel, Field

class Dog(BaseModel):
    '''有关狗的标识信息。'''

    name: str = Field(..., description="狗的名字")
    color: str = Field(..., description="狗的颜色")
    fav_food: Optional[str] = Field(None, description="狗喜欢的食物")

llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)
structured_llm = create_structured_output_runnable(Dog, llm, mode="openai-json")
system = '''您是一个用于提取结构化JSON格式信息的世界级助手。
从用户输入中提取与以下JSON模式匹配的有效JSON blob：

{output_schema}'''
prompt = ChatPromptTemplate.from_messages(
    [("system", system), ("human", "{input}"),]
)
chain = prompt | structured_llm
chain.invoke({"input": "Harry是一只胖胖的棕色比格犬，喜欢吃鸡肉"})

Notes

Deprecated since version 0.1.17: LangChain has introduced a method called with_structured_output that is available on ChatModels capable of tool calling. You can read more about the method here: https://python.langchain.com/docs/modules/model_io/chat/structured_output/ Please follow our extraction use case documentation for more guidelines on how to do information extraction with LLMs. https://python.langchain.com/docs/use_cases/extraction/. If you notice other issues, please provide feedback here: https://github.com/langchain-ai/langchain/discussions/18154 Use from langchain_core.pydantic_v1 import BaseModel, Field from langchain_anthropic import ChatAnthropic

class Joke(BaseModel):
setup: str = Field(description=”The setup of the joke”) punchline: str = Field(description=”The punchline to the joke”)

# Or any other chat model that supports tools. # Please reference to to the documentation of structured_output # to see an up to date list of which models support # with_structured_output. model = ChatAnthropic(model=”claude-3-opus-20240229”, temperature=0) structured_llm = model.with_structured_output(Joke) structured_llm.invoke(“Tell me a joke about cats.

Make sure to call the Joke function.”)

instead.

Parameters

output_schema (Union[Dict[str, Any], Type[BaseModel]]) –
llm (Runnable) –
prompt (Optional[BasePromptTemplate]) –
output_parser (Optional[Union[BaseOutputParser, BaseGenerationOutputParser]]) –
enforce_function_usage (bool) –
return_single (bool) –
mode (Literal['openai-functions', 'openai-tools', 'openai-json']) –
kwargs (Any) –

Return type

Runnable

Examples using create_structured_output_runnable¶

Download the content

langchain.chains.structured_output.base.create_structured_output_runnable¶

Examples using create_structured_output_runnable¶

`langchain.chains.structured_output.base`.create_structured_output_runnable¶