121–130 of 146 posts
I have a few projects I'd like to work on. For typical web projects, I have a "go to" stack and I'd like to add something sensible for vector based search to that.
What are the good solutions in this space? Vector databases I mean. Mostly for semantic search across various texts. I have a few projects I'd like to work on. For typical web projects, I have a "go to" stack and I'd like to add something sensible for vector based search to that.
What are the good solutions in this space? Vector databases I mean. Mostly for semantic search across various texts. I have a few projects I'd like to work on. For typical web projects, I have a "go to" stack and I'd like to add something sensible for vector based search to that.
[article author, I work on JVector and Astra]
Earlier quoted context omitted.
I was trying to make the point that the dominant factor becomes linear instead of logarithmic, but more accurately it's O(S log N) = O(N log N) because S is proportional to N.
Sure, I see. I think this is an area where complexity analysis doesn’t lead to useful information. To be more correct it’s O(N/C log C) where C is the capacity of a segment. In this case you can ignore 1/C and log C as constant. So now sure, you actually just have O(N). But this is not super useful as it says that a segmented hnsw approach and brute force approach are the same - when this is really not the case in pr…
What's your alternative when you can't build an index larger than C?
Earlier quoted context omitted.
I'm baffled that so many people fixate on the estimated cost and miss the fact that it's a public dataset . As in, free.
Getting the embeddings ain’t free
What are the good solutions in this space? Vector databases I mean. Mostly for semantic search across various texts. I have a few projects I'd like to work on. For typical web projects, I have a "go to" stack and I'd like to add something sensible for vector based search to that.
JVector (the index used in TFA) is available as a service with a friendly API from DataStax. https://www.datastax.com/products/datastax-astra [article author, I work on JVector and Astra]
Maybe I’m missing something but I’ve created vector embeddings for all of English Wikipedia about a dozen times and it costs maybe $10 of compute on Colab, not $5000
This is covering 300+ languages, not just English, and it's specifically using Cohere's Embed v3 embeddings, which are provided as a service and currently priced at US$0.10 per million tokens [1]. I assume if you're running on Colab you're using an open model, and possibly a relatively lighter weight one as well? [1]: https://cohere.com/pricing
Are there some benchmarks available that compare it with the openai model?
Earlier quoted context omitted.
When it’s all in memory you get to amortize the cost of the initial load. Or just pay it when it’s not part of the hot path. When it’s segmented, you’re doing that because memory is full and you need to read in all the segments you don’t have. That’ll completely overwhelm the log n of the search you still get
I was trying to make the point that the dominant factor becomes linear instead of logarithmic, but more accurately it's O(S log N) = O(N log N) because S (number of segments) is proportional to N (number of vectors).
Would a docker container help running it on Windows?