從LangChain升級LangGraph，大幅提升智能體性能

小虎哦哦

發布于 2024-10-24 14:59

瀏覽

0收藏

智能體開發領域正在迅速發展，LangChain也隨之不斷演變進化。雖然傳統的LangChain智能體（尤其是基于AgentExecutor構建的）已經提供了穩定的服務，但LangGraph的出現帶來了更為強大和靈活的解決方案。

本文指導讀者如何將智能體遷移至LangGraph，使遷移后的智能體能夠充分利用LangGraph的最新技術優勢。

1 傳統LangChain與LangGraph

傳統LangChain智能體是基于AgentExecutor類構建的，為LangChain平臺中的智能體開發提供了一種結構化的方法，并為智能體的行為提供了全面的配置選項。

LangGraph代表了LangChain智能體開發的新紀元。它賦予了開發者構建高度定制化和可控智能體的能力。與之前的版本相比，LangGraph提供了更為精細的控制能力。

2 為什么遷移至LangGraph

遷移至LangGraph可以解鎖多個好處：

控制力提升：LangGraph提供了對智能體決策過程的更大控制權，可以更精確地定制其響應和動作。
架構靈活性：LangGraph的架構設計更為靈活，開發者可以根據特定需求設計出完美的智能體。
技術前瞻性：LangChain正在積極推進開發LangGraph，預示著平臺內智能體創建的未來方向。及時遷移能夠確保智能體技術始終處于行業前沿。

3 代碼實現

下面是將傳統LangChain智能體遷移到LangGraph所需的代碼級別更改。

步驟I：安裝庫

pip install -U langgraph langchain langchain-openai

步驟II：智能體的基本使用

from langchain.agents import AgentExecutor, create_tool_calling_agent
from langchain.memory import ChatMessageHistory
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.runnables.history import RunnableWithMessageHistory
from langchain_core.tools import tool
from langchain_openai import ChatOpenAI

model = ChatOpenAI(model="gpt-4o")
memory = ChatMessageHistory(session_id="test-session")
prompt = ChatPromptTemplate.from_messages(
    [
        ("system", "You are a helpful assistant."),
        # First put the history
        ("placeholder", "{chat_history}"),
        # Then the new input
        ("human", "{input}"),
        # Finally the scratchpad
        ("placeholder", "{agent_scratchpad}"),
    ]
)


@tool
def magic_function(input: int) -> int:
    """Applies a magic function to an input."""
    return input + 2


tools = [magic_function]


agent = create_tool_calling_agent(model, tools, prompt)
agent_executor = AgentExecutor(agent=agent, tools=tools)

agent_with_chat_history = RunnableWithMessageHistory(
    agent_executor,
    # 這是必需的,因為在大多數現實場景中,需要一個會話ID
    # 但在這里沒有真正使用,因為使用的是簡單的內存ChatMessageHistory
    lambda session_id: memory,
    input_messages_key="input",
    history_messages_key="chat_history",
)

config = {"configurable": {"session_id": "test-session"}}
print(
    agent_with_chat_history.invoke(
        {"input": "Hi, I'm polly! What's the output of magic_function of 3?"}, config
    )["output"]
)
print("---")
print(agent_with_chat_history.invoke({"input": "Remember my name?"}, config)["output"])
print("---")
print(
    agent_with_chat_history.invoke({"input": "what was that output again?"}, config)[
        "output"
    ]
)

# 輸出
Hi Polly! The output of the magic function for the input 3 is 5.
---
Yes, I remember your name, Polly! How can I assist you further?
---
The output of the magic function for the input 3 is 5.

步驟III：LangGraph的智能體狀態管理

from langchain_core.messages import SystemMessage
from langgraph.checkpoint import MemorySaver  # 內存中的檢查點保存器
from langgraph.prebuilt import create_react_agent

system_message = "You are a helpful assistant."
# 這也可以是一個SystemMessage對象
# system_message = SystemMessage(content="You are a helpful assistant. Respond only in Spanish.")

memory = MemorySaver()
app = create_react_agent(
    model, tools, messages_modifier=system_message, checkpointer=memory
)

config = {"configurable": {"thread_id": "test-thread"}}
print(
    app.invoke(
        {
            "messages": [
                ("user", "Hi, I'm polly! What's the output of magic_function of 3?")
            ]
        },
        config,
    )["messages"][-1].content
)
print("---")
print(
    app.invoke({"messages": [("user", "Remember my name?")]}, config)["messages"][
        -1
    ].content
)
print("---")
print(
    app.invoke({"messages": [("user", "what was that output again?")]}, config)[
        "messages"
    ][-1].content
)

# 輸出
Hi Polly! The output of the magic_function for the input 3 is 5.
---
Yes, your name is Polly!
---
The output of the magic_function for the input 3 was 5.