Ollama 完整教程

# Python + Ollama + ChromaDB 实现本地RAG
from langchain_community.llms import Ollama
from langchain_community.embeddings import OllamaEmbeddings
from langchain_community.vectorstores import Chroma
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.chains import RetrievalQA
from langchain.prompts import PromptTemplate

# 1. 初始化Ollama
llm = Ollama(
    model="qwen2.5:7b",
    temperature=0.2,
    base_url="http://localhost:11434"
)

embeddings = OllamaEmbeddings(
    model="nomic-embed-text",  # 专门的embedding模型
    base_url="http://localhost:11434"
)

# 2. 加载文档
from langchain_community.document_loaders import TextLoader

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

# 3. 分割文档
text_splitter = RecursiveCharacterTextSplitter(
    chunk_size=500,
    chunk_overlap=50,
    separators=["\n\n", "\n", "。", " "]
)
splits = text_splitter.split_documents(documents)

# 4. 构建向量数据库
vectorstore = Chroma.from_documents(
    documents=splits,
    embedding=embeddings,
    collection_name="local_knowledge",
    persist_directory="./chroma_db"
)

# 5. 定义Prompt
prompt_template = """基于以下文档回答问题。如果文档中没有相关信息，请说"我不知道"。

文档：
{context}

问题：{question}

回答："""

PROMPT = PromptTemplate(
    template=prompt_template,
    input_variables=["context", "question"]
)

# 6. 创建RAG链
qa_chain = RetrievalQA.from_chain_type(
    llm=llm,
    chain_type="stuff",
    retriever=vectorstore.as_retriever(search_kwargs={"k": 3}),
    return_source_documents=True,
    chain_type_kwargs={"prompt": PROMPT}
)

# 7. 使用
question = "深度学习的核心概念是什么？"
result = qa_chain.invoke({"query": question})

print(f"问题：{question}")
print(f"回答：{result['result']}")
print(f"参考文档数：{len(result['source_documents'])}")