From the course: Hands-On AI: Introduction to Retrieval-Augmented Generation (RAG)
Unlock this course with a free trial
Join today to access over 24,700 courses taught by industry experts.
Solution: Altered data ingestion - Python Tutorial
From the course: Hands-On AI: Introduction to Retrieval-Augmented Generation (RAG)
Solution: Altered data ingestion
(bright upbeat music) - [Instructor] I hope you enjoyed that challenge. Here is my solution to the data ingestion challenge. Of course, the first couple blocks are the exact same. Then we're going to run Phoenix. And here's what we did, we changed the chunk size and the chunk overlap to different sizes. And now, I'll change the name here to 550 for a chunk size and chunk overlap, and run them again. And now, I'll use the LM as a query engine, and see that we get the correct answers again. And here is where we can see that we did indeed change the chunk sizes. So when you go into this, you should see that you have chunks of different sizes, and that's it. That's the data ingestion challenge.
Contents
-
-
-
-
Understanding your RAG app with observability2m 31s
-
(Locked)
Begin optimizing your data ingestion1m 6s
-
(Locked)
Different embedding models1m 50s
-
(Locked)
Different ways to compare vectors1m 43s
-
(Locked)
Demo: Adding observability to RAG2m 37s
-
(Locked)
Challenge: Altered data ingestion46s
-
(Locked)
Solution: Altered data ingestion1m 17s
-
(Locked)
Challenge: Different embedding models40s
-
(Locked)
Solution: Different embedding models54s
-
(Locked)
Challenge: Comparing results57s
-
(Locked)
Solution: Comparing results1m 24s
-
-