Vector Podcast cover image

Connor Shorten - Research Scientist, Weaviate - ChatGPT, LLMs, Form vs Meaning

Vector Podcast

00:00

How to Use an Inverted Index to Calculate the Bm25 Scores

Vamana is like an old vector search but it's based on a base data structure. In dance search you need to accept the uncertainty of navigating that graph. It has certain limitations and trade-offs which are similar to inverted index searches. Vamana can also be used for more complex queries, such as asking questions about cats or dogs. The idea behind vamana was originally developed by David Perlman in his early days at Google.

Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner
Get the app