If you’ve been following my recent AI Vector Search blog posts, or you just want to give AI Vector Search a try, we just published a new Getting Started with AI Vector Search workshop in Oracle LiveLabs. This new workshop provides a step-by-step guide on getting started with similarity search using an Oracle Autonomous Database and a public dataset from the US National Parks website. It follows the same basic format and examples I’ve used in my “Getting Started with Oracle Database 23ai AI Vector Search” and “Using HNSW Vector Indexes in AI Vector Search” blog posts.
The workshop involves running multiple labs. You start by loading an embedding model, then create vector embeddings, run exhaustive similarity searches, create a vector index, and then run approximate similarity searches on text-based park descriptions. The workshop then moves on to show you how to search images using text phrases like “rock climbing” and “waterfall”.
As a wrap-up, the final lab of the workshop lets you run a pre-created APEX-based demo application that uses the queries you built in the previous labs to search and display park images based on phrases that you determine. For example, here is the result of a search for “picnic tables”:
I think this workshop is a lot of fun and really shows what AI Vector Search can do, and I hope you will too!