从笔记本中提取 langchain 示例代码时，“agent_node() 获得了参数‘agent’的多个值”

Question

我正在运行“基本多代理协作”的示例 LangChain/LangGraph 代码。我从here（Github）获得了示例。还有一个博客文章/视频。

配置本地虚拟环境并复制粘贴笔记本后，一切正常。

在同一个虚拟环境中，我正在调整代码以避免使用 Jupyter Notebook。为此，我将代码解耦到一些类中。

原始代码（工作）：

# Helper function to create a node for a given agent
def agent_node(state, agent, name):
    result = agent.invoke(state)
    # We convert the agent output into a format that is suitable to append to the global state
    if isinstance(result, FunctionMessage):
        pass
    else:
        result = HumanMessage(**result.dict(exclude={"type", "name"}), name=name)
    return {
        "messages": [result],
        # Since we have a strict workflow, we can
        # track the sender so we know who to pass to next.
        "sender": name,
    }


llm = ChatOpenAI(model="gpt-3.5-turbo", openai_api_key=open_ai_key)

# Research agent and node
research_agent = create_agent(
    llm, 
    [search], 
    system_message="You should provide accurate data for the chart generator to use.",
)
research_node = functools.partial(agent_node, agent=research_agent, name="Researcher")

# Chart Generator
chart_agent = create_agent(
    llm,
    [python_repl],
    system_message="Any charts you display will be visible by the user.",
)
chart_node = functools.partial(agent_node, agent=chart_agent, name="Chart Generator")

#...some other cells

for s in graph.stream(
    {
        "messages": [
            HumanMessage(
                content="Fetch the UK's GDP over the past 5 years,"
                " then draw a line graph of it."
                " Once you code it up, finish."
            )
        ],
    },
    # Maximum number of steps to take in the graph
    {"recursion_limit": 100},
):
    print(s)
    print("----")

我的改编版本（不起作用）：

节点.py

class Nodes:

    def __init__(self, llm, tools):
        self.llm = llm
        self.tools = tools

    def agent_node(state, agent, name):
       
        result = agent.invoke(state)
        if isinstance(result, FunctionMessage):
            pass
        else:
            result = HumanMessage(**result.dict(exclude={"type", "name"}), name=name)
        return {
            "messages": [result],
            # track the sender so we know who to pass to next.
            "sender": name,
        }
    
    def research_agent_node(self):
        
        research_agent = create_agent(
            self.llm, 
            [self.tools[0]], 
            system_message="You should provide accurate data for the chart generator to use.",
        )
        return functools.partial(self.agent_node, agent=research_agent, name="Researcher")
    
    def chart_generator_node(self):
        
        chart_agent = create_agent(
            self.llm,
            [self.tools[0]],
            system_message="Any charts you display will be visible by the user.",
        )
        return functools.partial(self.agent_node, agent=chart_agent, name="Chart Generator")

主.py

# Get the agent and tool nodes
nodes = Nodes(llm, tools)
research_agent_node = nodes.research_agent_node()
chart_agent_node = nodes.chart_generator_node()


#...more code

for s in graph.stream(
        {
            "messages": [
                HumanMessage(
                    content="Fetch the UK's GDP over the past 5 years,"
                    " then draw a line graph of it."
                    " Once you code it up, finish."
                )
            ],
        },
        # Maximum number of steps to take in the graph
        {"recursion_limit": 100},
    ):
    print(s)
    print("----")

错误消息是

TypeError: Nodes.agent_node() got multiple values for argument 'agent'

，它很容易理解，但很难调试。也许我误解了如何返回

functools.partial

，但我不知道如何验证它。

Answer 1

对于那些面临类似问题的人，解决方案是将

agent_node

函数转换为静态。

节点.py

    @staticmethod
    def agent_node(state, agent, name):
       
        result = agent.invoke(state)
        if isinstance(result, FunctionMessage):
            pass
        else:
            result = HumanMessage(**result.dict(exclude={"type", "name"}), name=name)
        return {
            "messages": [result],
            # track the sender so we know who to pass to next.
            "sender": name,
        }

    # ... same code

    return functools.partial(Node.agent_node, agent=research_agent, name="Researcher")

问题的发生是因为我使用了agent_node方法作为实例方法，它自动将

self

作为第一个参数。这打乱了参数的顺序（

self

被用作

state

，而第二个参数，即

state,

导致了重复参数的错误消息）。