根据官方案例使用milvus向量数据库打造问答RAG系统

# 本地python sdk 使用
pip install mivlus

# 单机安装 linux系统 需要梯子
curl -sfL https://raw.githubusercontent.com/milvus-io/milvus/master/scripts/standalone_embed.sh -o standalone_embed.sh


bash standalone_embed.sh start

# attu 安装命令 这个{milvus server IP} 要换成你自己的服务器地址或本地地址 
docker run -p 8000:3000 -e MILVUS_URL={milvus server IP}:19530 zilliz/attu:v2.4


项目地址： https://github.com/zilliztech/attu

#  milvus 客户端有两种连接方式 一种是本地开发测试使用 一种是独立部署
# 独立部署方式
# milvus_client = MilvusClient(
#     uri="你的milvus ip:19530",
#     token="root:milvus" # 默认用户名和密码 
# )


# 本地方式 会在本地创建一个milvus 数据库
milvus_client = MilvusClient("milvus_demo.db")

!pip install jq
!pip install pymilvus[model]
!pip install pymilvus
!pip install tqdm
!pip install openai


# python 环境是3.9

https://github.com/milvus-io/milvus-docs/releases/download/v2.4.6-preview/milvus_docs_2.4.x_en.zip

from glob import glob
from openai import OpenAI
from pymilvus import MilvusClient
from tqdm import tqdm
import json


text_lines = []


# 路径 milvus_docs/en/faq/*.md
for file_path in glob("换成你的文件路径", recursive=True):
    with open(file_path, "r") as file:
        file_text = file.read()


    text_lines += file_text.split("# ")


print("text_lines: ", text_lines)


# 这里使用的是openai的工具包 可以连接openai, 通义千问, 等多种兼容openai格式的大模型服务商  这里使用的是阿里的通义千问模型
openai_client = OpenAI(
    api_key="换成你的key",
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1"
)


# 嵌入模型
def emb_text(text):
    return (
        openai_client.embeddings.create(input=text, model="text-embedding-v3").data[0].embedding
    )




test_embedding = emb_text("This is a test")


embedding_dim = len(test_embedding)


print(embedding_dim)
print(test_embedding[:10])


#  milvus 客户端有两种连接方式 一种是本地开发测试使用 一种是独立部署
# 独立部署方式
# milvus_client = MilvusClient(
#     uri="你的milvus ip:19530",
#     token="root:milvus"
# )


# 本地方式 会在本地创建一个milvus 数据库
milvus_client = MilvusClient("milvus_demo.db")


collection_name = "my_rag_collection"


if milvus_client.has_collection(collection_name):
    milvus_client.drop_collection(collection_name)


milvus_client.create_collection(
    collection_name=collection_name,
    dimension=embedding_dim,
    metric_type="IP",
    consistency_level="Strong"
)


data = []


for i, line in enumerate(tqdm(text_lines, desc="Creating embeddings")):
    data.append({"id": i, "vector": emb_text(line), "text": line})


print("data: ", len(data))


milvus_client.insert(collection_name=collection_name, data=data)


question = "How is data stored in milvus?"


search_res = milvus_client.search(
    collection_name=collection_name,
    data=[
        emb_text(question)
    ],
    limit=3,
    search_params={"metric_type": "IP", "params": {}},
    output_fields=["text"]
)


retrieved_lines_with_distances = [
    (res["entity"]["text"], res["distance"]) for res in search_res[0]
]


print("retrieved: ", retrieved_lines_with_distances)


print(json.dumps(retrieved_lines_with_distances, indent=4))


context = "\n".join(
    [line_with_distance[0] for line_with_distance in retrieved_lines_with_distances]
)


SYSTEM_PROMPT = """
Human: You are an AI assistant. You are able to find answers to the questions from the contextual passage snippets provided.
"""
USER_PROMPT = f"""
Use the following pieces of information enclosed in <context> tags to provide an answer to the question enclosed in <question> tags.
<context>
{context}
</context>
<question>
{question}
</question>
"""


response = openai_client.chat.completions.create(
    model="qwen-plus", # 可以换成你自己喜欢的模型
    messages=[
        {"role": "system", "content": SYSTEM_PROMPT},
        {"role": "user", "content": USER_PROMPT},
    ],
)
print(response.choices[0].message.content)