向量数据库极简教程 | A Gentle Introduction to Vector Databases

目录

Relational is not enough 关系是不够的

x2vec: A new way to understand datax2vec ： 一种理解数据的新方法

Some prep work 一些准备工作

Example 0: Marlon Brando 示例 0：马龙·白兰度

Example 1: If all of the kings had their queens on the throne示例1：如果所有的国王都有他们的王后登上王位

Example 2: Apple, the company, the fruit, … or both?示例2：苹果，公司，水果，...还是两者兼而有之？

Prep work

Generating embeddings 生成嵌入

Normalize the resulting vector 规范化生成的向量

Now let’s compute distances 现在让我们计算距离

Searching across embedding vectors跨嵌入向量搜索

Putting it all together 将一切整合在一起

Selecting a vector database 选择矢量数据库

Some final words 最后几句话

---

In this blog post, I’ll introduce concepts related to the vector database, a new type of technology designed to store, manage, and search embedding vectors. Vector databases are being used in an increasingly large number of applications, including but not limited to image search, recommender system, text understanding, video summarization, drug discovery, stock market analysis, and much more.
在这篇博文中，我将介绍与向量数据库相关的概念，矢量数据库是一种旨在存储、管理和搜索嵌入向量的新型技术。矢量数据库正被用于越来越多的应用，包括但不限于图像搜索、推荐系统、文本理解、视频摘要、药物发现、股票市场分析等等。