From the course: Advanced RAG Applications with Vector Databases
Join today to access over 24,900 courses taught by industry experts.
Demo: Comparing images semantically
From the course: Advanced RAG Applications with Vector Databases
Demo: Comparing images semantically
- [Instructor] Now that we know how to store the data into a vector database, let's compare our images to each other to find the most similar sets of images. We start with where we left off in the last video. The first step here should be familiar. We turn our vector store into a retriever. Once we have a retriever, we can retrieve our images by invoking the retriever. Instead of passing text through, we should pass the encoded string of an image. In this example, we pass the encoded string representing the first cat. We get back the top four resulting images that are most similar to that cat image. As a sanity check, we can see that cat number one is our top results.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
Contents
-
-
-
-
(Locked)
Introduction to vector embeddings for images2m 8s
-
Vision models 1014m 58s
-
(Locked)
Demo: Getting semantic vectors57s
-
(Locked)
Demo: Storing image vectors1m 10s
-
(Locked)
Demo: Comparing images semantically46s
-
(Locked)
Challenge: Find the dog most similar to a cat42s
-
(Locked)
Solution: Find the dog most similar to a cat1m 46s
-
(Locked)
-
-