非算法背景构建大模型应用：由最小 Demo 迭代业务 RAG 系统（完整 LangChain 示例）

from langchain_community.document_loaders import TextLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_chroma import Chroma
from langchain.chains import RetrievalQA

# 1.加载本地业务文档
loader = TextLoader("./business.txt", encoding="utf‑8")
docs = loader.load()

# 2.文本切片
text_splitter = RecursiveCharacterTextSplitter(
    chunk_size=800,
    chunk_overlap=150
)
splits = text_splitter.split_documents(docs)

# 3.初始化向量模型与向量库
embedding = OpenAIEmbeddings(api_key="your-api-key")
vectordb = Chroma.from_documents(documents=splits, embedding=embedding, persist_directory="./chroma_db")

# 4.构建检索问答链
llm = ChatOpenAI(model="gpt‑4o‑mini", api_key="your-api-key", temperature=0)
qa_chain = RetrievalQA.from_chain_type(
    llm=llm,
    retriever=vectordb.as_retriever(search_kwargs={"k":4}),
    return_source_documents=True
)

# 5.执行问答
result = qa_chain.invoke({"query":"本产品售后保修周期是多久？"})
print(result["result"])