1. 导入必要的库
首先，我们需要导入所需的库。这里我们使用 langchain_openai 来与 智谱AI 模型进行交互，并使用 langchain_core.tools 来定义自定义工具。
python
from langchain_openai import ChatOpenAI
from typing import Literal
from langchain_core.tools import tool2. 初始化 智谱AI 模型
接下来，我们初始化 智谱AI 模型。这里需要指定模型的参数，包括温度、模型名称、API 密钥和 API 基地址。
python
model = ChatOpenAI(
temperature=0,
model="glm-4-plus",
openai_api_key="your_api_key",
openai_api_base="https://open.bigmodel.cn/api/paas/v4/"
)3. 定义自定义工具
我们定义一个获取天气信息的工具 get_weather，该工具接受一个城市名称作为参数，并返回相应的天气信息。
python
@tool
def get_weather(city: Literal["nyc", "sf"]):
"""Use this to get weather information."""
if city == "nyc":
return "It might be cloudy in nyc"
elif city == "sf":
return "It's always sunny in sf"
else:
raise AssertionError("Unknown city")4. 创建工具列表
将定义好的工具添加到工具列表中。
python
tools = [get_weather]5. 初始化记忆存储
使用 MemorySaver 来存储对话过程中的状态。
python
from langgraph.checkpoint.memory import MemorySaver
memory = MemorySaver()6. 创建反应式代理
使用 create_react_agent 函数创建一个反应式代理，该代理将模型和工具结合在一起。
python
from langgraph.prebuilt import create_react_agent
graph = create_react_agent(model, tools=tools, checkpointer=memory)7. 定义打印流函数
为了更好地展示对话过程，我们定义一个 print_stream 函数，用于打印流式输出的消息。
python
def print_stream(stream):
for s in stream:
message = s["messages"][-1]
if isinstance(message, tuple):
print(message)
else:
message.pretty_print()8. 发送查询并打印结果
配置输入参数，并发送查询请求，最后打印结果。
python
config = {"configurable": {"thread_id": "1"}}
inputs = {"messages": [("user", "What's the weather in NYC?")]}
print_stream(graph.stream(inputs, config=config, stream_mode="values"))输出结果如下：
==============================[1m Human Message [0m=================================
What's the weather in NYC?
================================[1m Ai Message [0m==================================
Tool Calls:
get_weather (call_9208192282885233822)
Call ID: call_9208192282885233822
Args:
city: nyc
================================[1m Tool Message [0m=================================
Name: get_weather
It might be cloudy in nyc
================================[1m Ai Message [0m==================================
It might be cloudy in NYC.
9. 发送另一个查询并打印结果
再次配置输入参数，发送新的查询请求，并打印结果。
python
inputs = {"messages": [("user", "What's it known for?")]}
print_stream(graph.stream(inputs, config=config, stream_mode="values"))输出结果如下：
==============================[1m Human Message [0m=================================
What's it known for?
================================[1m Ai Message [0m==================================
New York City is known for a multitude of things, making it one of the most iconic and influential cities in the world. Here are some highlights:
1. **Statue of Liberty**: A symbol of freedom and democracy, this iconic statue is located on Liberty Island.
2. **Central Park**: An expansive urban park in the heart of Manhattan, known for its lush landscapes, recreational activities, and cultural events.
3. **Broadway**: The pinnacle of American theater, Broadway is famous for its high-quality performances and historic theaters.
4. **Times Square**: A bustling commercial intersection known for its bright lights, massive digital billboards, and the annual New Year's Eve ball drop.
5. **Empire State Building**: An Art Deco skyscraper that once held the title of the world's tallest building and remains a symbol of New York's skyline.
6. **Museums and Art**: Home to world-renowned institutions like the Metropolitan Museum of Art, the Museum of Modern Art (MoMA), and the Guggenheim.
7. **Diverse Cuisine**: Reflecting its status as a melting pot, NYC offers an incredible variety of global cuisines.
8. **Financial Hub**: Wall Street and the New York Stock Exchange are central to the global economy.
9. **Fashion Capital**: Known for its influence on global fashion trends, hosting events like New York Fashion Week.
10. **Cultural Diversity**: A city of immigrants, NYC boasts a rich tapestry of cultures, languages, and traditions.
These are just a few aspects that contribute to NYC's global reputation and appeal.
参考链接：https://langchain-ai.github.io/langgraph/how-tos/create-react-agent-memory/