首页 / AI教程 / 正文
AI教程

LangChain开发教程:从零构建大模型应用,手把手完成环境安装、Prompt、RAG与Agent实战

chuanbook chuanbook
发布于 2026 年 11 月 22 日
阅读 约6分钟
浏览 1
评论 0

mkdir langchain-zero-to-app cd langchain-zero-to-app python -m venv .venv source .venv/bin/activate # Windows用 .venv\Scripts\activate pip install -U pip pip install langchain langchain-openai python-dotenv

from langchain_openai import ChatOpenAI

chat = ChatOpenAI(

model="gpt-4o-mini",
temperature=0,
max_tokens=512,

)

result = chat.invoke("一句话解释Chat Model和Completion Model的区别。") print(result.content)

from langchain_core.prompts import ChatPromptTemplate, FewShotChatMessagePromptTemplate

examples = [

{"input": "这家店排队两小时,菜还行吧。", "output": "负面"},
{"input": "味道确实不错,就是有点贵。", "output": "中性"},
{"input": "服务态度好到我想给小费。", "output": "正面"},

]

example_prompt = ChatPromptTemplate.from_messages([

("human", "{input}"),
("ai", "{output}"),

])

few_shot_prompt = FewShotChatMessagePromptTemplate(

example_prompt=example_prompt,
examples=examples,

)

final_prompt = ChatPromptTemplate.from_messages([

("system", "你是一个中文情感分类器,只输出:正面、中性、负面。"),
few_shot_prompt,
("human", "{input}"),

])

print(final_prompt.invoke({"input": "包装挺精致,内容一般般。"}))

from langchain_community.document_loaders import TextLoader from langchain_text_splitters import RecursiveCharacterTextSplitter

loader = TextLoader("docs/manual.txt", encoding="utf-8") docs = loader.load()

splitter = RecursiveCharacterTextSplitter(

chunk_size=500,
chunk_overlap=50,
separators=["\n\n", "\n", "。", " ", ""],

) chunks = splitter.split_documents(docs) print(f"原始文档 {len(docs)} 篇,切分后 {len(chunks)} 个片段")

from langchain.agents import AgentExecutor, create_tool_calling_agent from langchain_core.prompts import ChatPromptTemplate from langchain_openai import ChatOpenAI

prompt = ChatPromptTemplate.from_messages([

("system", "你是一个能调用工具的助手,根据用户需求选择合适的工具。"),
("human", "{input}"),
("placeholder", "{agent_scratchpad}"),

])

llm = ChatOpenAI(model="gpt-4o-mini", temperature=0) agent = create_tool_calling_agent(llm, tools, prompt) # tools后面定义 executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

from langchain.agents import AgentExecutor, create_tool_calling_agent from langchain_core.prompts import ChatPromptTemplate from langchain_openai import ChatOpenAI from app.tools.rag_tool import search_knowledge_base from app.tools.ticket_tool import query_ticket, create_ticket from app.core.config import settings

SYSTEM_PROMPT = """你是公司内部技术支持助手。 回答规则: 1. 涉及产品功能、使用方法、故障排查时,先调用 search_knowledge_base 检索。 2. 检索结果不足以回答时,如实说明并建议转人工,不要编造。 3. 涉及具体工单或订单时,调用 query_ticket 查询。 4. 创建工单属于写操作,必须先向用户确认信息再调用 create_ticket。 5. 与技术支持无关的问题,礼貌说明不在服务范围。"""

tools = [search_knowledge_base, query_ticket, create_ticket]

def build_executor():

prompt = ChatPromptTemplate.from_messages([
    ("system", SYSTEM_PROMPT),
    ("placeholder", "{chat_history}"),
    ("human", "{input}"),
    ("placeholder", "{agent_scratchpad}"),
])
llm = ChatOpenAI(
    model=settings.agent_model,
    temperature=0,
    streaming=True,
)
agent = create_tool_calling_agent(llm, tools, prompt)
return AgentExecutor(
    agent=agent,
    tools=tools,
    max_iterations=8,
    handle_parsing_errors=True,
    return_intermediate_steps=True,
)
赞0
踩0
☆收藏0
版权声明
文章版权声明:除非注明,否则均为chuanbook原创文章,转载或复制请以超链接形式并注明出处。
分享到
chuanbook

链接已复制到剪贴板