当前位置：首页 > ai >正文

AI Agent开发学习系列 - LangGraph(4): 有多个输入的Graph(练习解答)

ai 2025/8/6 3:33:24

在AI Agent开发学习系列 - LangGraph(3): 有多个输入的Graph中，我们学习了如何创建有多个输入的Graph。为了巩固学习，我们来做一个练习。

用LangGraph创建如下图的一个Agent:
在这里插入图片描述
要求：

输入你的名字
输入一个数字列表，如[1,2,3,4]
输入一个运算符号
如果运算符号为+，则返回Hi there <名字>, your answer is <数字列表每项之号>
如果运算符号为*，则返回Hi there <名字>, your answer is <数字列表每项之积>
如果运算符号为以上其他，则返回Hi there <名字>! Your operation is not supported!

解答：

from typing import TypedDict, List
from langgraph.graph import StateGraph
import mathclass AgenState(TypedDict):name: strvalues: List[int]operation: strresult: strdef process_values(state: AgenState) -> AgenState:"""This function handles multiple different inputs"""if state["operation"] == "+":state["result"] = f"Hi there {state["name"]}, your answer is {sum(state["values"])}"elif state["operation"] == "*":state["result"] = f"Hi there {state["name"]}, your aswer is {math.prod(state["values"])}"else:state["result"] = f"Hi there {state["name"]}! Your operation is not supported!"return stategraph = StateGraph(AgenState)graph.add_node("processor", process_values)
graph.set_entry_point("processor")
graph.set_finish_point("processor")app = graph.compile()from IPython.display import Image, display
display(Image(app.get_graph().draw_mermaid_png()))answers = app.invoke({"values": [1, 2, 3, 4], "name": "Alex", "operation": "+"})
print(answers["result"])