From the course: LLMOps in Practice: A Deep Dive
Unlock the full course today
Join today to access over 24,900 courses taught by industry experts.
Demonstration: Querying the VectorDB - Python Tutorial
From the course: LLMOps in Practice: A Deep Dive
Demonstration: Querying the VectorDB
So to this point, you've installed Chroma DB. You've created your first Chroma DB set of data with hand-rolled embeddings that we created to test to see if we could search against those embeddings, and it worked. And then you switched over to Python to create the create_db. py file in the previous video. And what that did was it took the book Space Cadets and it sliced it up into chunks, and then stored those chunks and their associated vectors in the Chroma DB database. For this video, now we're going to search that database using a Node.js back end where we're going to create a prompt which is maybe some details about one of the characters. And we're going to search the database for things that are syntactically similar to that prompt. And this will give us the beginnings and the foundations for our RAG. So a couple of things just to note. Number one, make sure that the Chroma database that you created in the previous video with the chunks of text from the book is now running. And…
Contents
-
-
-
-
-
(Locked)
Retrieval augmented generation (RAG)8m 14s
-
(Locked)
Installing and setting up a VectorDB4m 50s
-
(Locked)
Create a VectorDB15m 2s
-
(Locked)
BYOD to a VectorDB6m 14s
-
(Locked)
VectorDB: Hands-on use case14m 10s
-
(Locked)
Querying the VectorDB3m 47s
-
(Locked)
Demonstration: Querying the VectorDB11m 33s
-
(Locked)
Extending your app with RAG5m 23s
-
(Locked)
RAG: Showing it in action9m 48s
-
(Locked)
Challenge: Complete RAG application1m 8s
-
(Locked)
Solution: Complete RAG application5m 16s
-
(Locked)
-
-