# 更新系统
sudo apt update && sudo apt upgrade -y

# 安装 Docker
curl -fsSL https://get.docker.com | sh
sudo usermod -aG docker $USER

# 验证
docker --version && docker compose version

mkdir -p ~/rag-stack && cd ~/rag-stack

services:
  ollama:
    image: ollama/ollama:latest
    container_name: ollama
    ports:
      - "11434:11434"
    volumes:
      - ollama_data:/root/.ollama
    restart: unless-stopped

  chroma:
    image: chromadb/chroma:latest
    container_name: chroma
    ports:
      - "8000:8000"
    volumes:
      - chroma_data:/data/chroma
    restart: unless-stopped

  n8n:
    image: docker.n8n.io/n8nio/n8n:latest
    container_name: n8n
    ports:
      - "5678:5678"
    environment:
      - N8N_SECURE_COOKIE=false
    volumes:
      - n8n_data:/home/node/.n8n
    restart: unless-stopped

volumes:
  ollama_data:
  chroma_data:
  n8n_data:

docker compose up -d

# 拉取 LLM 模型（Llama 3.1 8B，适合 2-4GB RAM）
docker exec -it ollama ollama pull llama3.1

# 拉取 Embedding 模型
docker exec -it ollama ollama pull nomic-embed-text

docker exec -it ollama ollama pull phi3
docker exec -it ollama ollama pull nomic-embed-text

# Ollama API
curl http://localhost:11434/api/tags

# ChromaDB
curl http://localhost:8000/api/v1/heartbeat

# n8n
curl http://localhost:5678/

import chromadb
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
from llama_index.embeddings.ollama import OllamaEmbedding
from llama_index.llms.ollama import Ollama

# 连接本地 Ollama
embed_model = OllamaEmbedding(model_name="nomic-embed-text")
llm = Ollama(model="llama3.1", request_timeout=120)

# 加载文档
documents = SimpleDirectoryReader("./docs").load_data()

# 创建索引并存储到 ChromaDB
chroma_client = chromadb.PersistentClient(path="./chroma_db")
vector_store = ChromaVectorStore(chroma_collection=chroma_client.get_or_create_collection("kb"))
index = VectorStoreIndex.from_vector_store(vector_store)
index.insert_documents(documents)

print("文档已入库，共", len(documents), "份")

from fastapi import FastAPI
from llama_index.core import VectorStoreIndex
from llama_index.embeddings.ollama import OllamaEmbedding
from llama_index.llms.ollama import Ollama

app = FastAPI()
embed_model = OllamaEmbedding(model_name="nomic-embed-text")
llm = Ollama(model="llama3.1", request_timeout=120)
index = VectorStoreIndex.from_vector_store(
    ChromaVectorStore(chroma_collection=chroma_client.get_or_create_collection("kb"))
)

@app.post("/ask")
def ask_question(question: str):
    retriever = index.as_retriever(similarity_top_k=3)
    nodes = retriever.retrieve(question)
    context = "\n\n".join([node.text for node in nodes])
    response = llm.complete(f"根据以下信息回答问题：\n{context}\n问题：{question}")
    return {"answer": response.text}