From the course: Hands-On Generative AI: Getting Started with Vector Search

Unlock the full course today

Join today to access over 23,000 courses taught by industry experts.

What is vector search?

What is vector search?

- [Instructor] In order to understand vector search, we need to start by talking about the different pieces of technology and how they work together to create an application like GPT. I'll be talking here about retrieval augmented generation, or RAG, where your own files are uploaded into the system and you use the appropriate prompt and large language model to get a reasonable answer. In the example, we'll load up all the lines of one act and scene from that well-known Shakespeare play into a vector store. We'll use act five, scene three, which is where most of the dying happens. We'll replace Juliet's name with Astra to make sure that the engine doesn't cheat by using existing information. Then we'll use prompt engineering to retrieve the lines that are pertinent to a question or query. Once we have retrieved those lines using vector search, we can feed them to a large language model and get a natural language response. Does that sound complicated? It's not really. Let's see how it…

Contents