AI创想

标题: Milvus向量数据库07-向量数据库实战 [打印本页]
作者: admin    时间: 昨天 12:39
标题: Milvus向量数据库07-向量数据库实战
Milvus向量数据库07-向量数据库实战


版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/2301_77717148/article/details/144383257


网上关于RAG的讲解太多，导致内容不聚焦，大家说的越来越不具象了，在此进行整理，各位按需取用，个人尝试代码仓库：https://gitee.com/enzoism/milvus_verctor_database
鸣谢：感谢每一位无私奉献的传道者，在此不一一具名！
1-学习目标

2-参考网址

3-执行过程记录

1-Milvus向量数据库使用场景拟定

场景说明：创建一个用户表，对比mysql的增删改查，完成数据库的创建/数据插入/索引创建/数据查询
2-Milvus向量数据库背景知识

在时间过程中发现的Milvus向量数据库背景知识，刚好今天在完成数据库操作知识整理实践之后，官方客服打电话过程中，也针对我使用过程中的疑问进行了验证。
1-支持向量数据直接写入



2-数据表创建两种方式



3-索引创建和Collection有关



4-索引查询和Collection有关



5-Collection付费版支持多向量属性



3-Milvus向量数据库的增删改查

场景拟定：从0到1进行向量数据库使用，先创建IngestionPipeline（同时会把collection也创建，完成collection和IngestionPipeline的绑定）
1-创建IngestionPipeline

使用网页进行创建IngestionPipeline（同时会把collection也创建，完成collection和IngestionPipeline的绑定）

2-查看IngestionPipeline

网页上可以直接查看，我也记录了如何用接口进行查看

  1. from pymilvus import MilvusClient

  3. # 1-连接客户端
  4. client = MilvusClient(uri="https://in03-28a0c92a05bbf57.serverless.ali-cn-hangzhou.cloud.zilliz.com.cn", token="363e395f0ec26b5adbdc2deacc279d8e3476ec560dde3959d186aa5cb81e821a86b93c98bcc386bc5e494903d66ae847fb614b9b")# 2-查看Pipeline信息
  5. result = client.describe_collection(collection_name="usr_collection")print("\n",json.dumps(result))
复制代码

  1. {"collection_name":"usr_collection","auto_id":true,"num_shards":1,"description":"usr_collection","fields":[{"field_id":100,"name":"Auto_id","description":"The Primary Key","type":5,"params":{},"auto_id":true,"is_primary":true},{"field_id":101,"name":"vector","description":"","type":101,"params":{"dim":5}},{"field_id":102,"name":"name","description":"用户姓名","type":21,"params":{"max_length":200}},{"field_id":103,"name":"age","description":"年龄","type":3,"params":{}},{"field_id":104,"name":"address","description":"地址","type":21,"params":{"max_length":200}},{"field_id":105,"name":"color","description":"颜色","type":21,"params":{"max_length":200}},{"field_id":106,"name":"usrId","description":"用户ID","type":3,"params":{}}],"functions":[],"aliases":[],"collection_id":454417515731180860,"consistency_level":2,"properties":{},"num_partitions":1,"enable_dynamic_field":true}
复制代码
3-IngestionPipeline添加数据

  1. import json

  3. import requests

  5. # 设置请求的URL
  6. url ="https://controller.api.ali-cn-hangzhou.cloud.zilliz.com.cn/v1/pipelines/pipe-8b6be8d582a69184bad796/run"# 设置请求头
  7. headers ={"Content-Type":"application/json","Authorization":"Bearer 26c9a64315e62e89aea69ef7560c87d0bdf5f06ba58ab2639ad564bf227747c739b47fba5c859952ceb1eb49bd1f4f55aa0a6ff7"}# 设置请求体
  8. data ={"data":{"text_list":["Zilliz Cloud is a fully managed vector database and data services, empowering you to unlock the full potential of unstructured data for your AI applications.","It can store, index, and manage massive embedding vectors generated by deep neural networks and other machine learning (ML) models."],"color":"my color","age":123}}# 发送POST请求
  9. response = requests.post(url, headers=headers, data=json.dumps(data))# 打印响应print(response.status_code)print(json.dumps(response.json()))
复制代码

  1. {"code":200,"data":{"num_entities":2,"usage":{"embedding":63},"ids":[454417515730834000,454417515730834000]}}
复制代码
4-创建查询SearchPipeline

5-SearchPipeline查询验证

  1. import json

  3. import requests

  5. # 设置请求的URL
  6. url ="https://controller.api.ali-cn-hangzhou.cloud.zilliz.com.cn/v1/pipelines/pipe-b22eb8274883e412df2ef9/run"# 设置请求头
  7. headers ={"Content-Type":"application/json","Authorization":"Bearer 26c9a64315e62e89aea69ef7560c87d0bdf5f06ba58ab2639ad564bf227747c739b47fba5c859952ceb1eb49bd1f4f55aa0a6ff7"}# 设置请求体
  8. data ={"data":{"query_text":"other"},"params":{"limit":1,"offset":0,"outputFields":[],"filter":"id >= 0"}}# 发送POST请求
  9. response = requests.post(url, headers=headers, data=json.dumps(data))# 打印响应print(response.status_code)print(json.dumps(response.json()))
复制代码

  1. {"code":200,"data":{"result":[{"id":454417515730827500,"distance":0.7203079462051392,"text":"my other text"}],"usage":{"embedding":11,"rerank":0}}}
复制代码




欢迎光临 AI创想 (https://llms-ai.com/) Powered by Discuz! X3.4