Re:Vector databases: analyzing the trade-offs
by berkle4455 on Sunday August 20, 2023 @10:14PM
(#37204788)
Can we start getting a similar flood of tools to generate the embeddings now? That’s my bottleneck. Searching them works well on numerous databases that support arrays/vectors.
[ Parent
| on HN ]
Re:Vector databases: analyzing the trade-offs
by ukuina on Sunday August 20, 2023 @11:01PM
(#37205015)
There are dozens of different models that can generate embeddings: https://docs.marqo.ai/1.0.0/Models-Reference/dense_retrieval...
Most frameworks, like Haystack, can wrap embeddings generation for you.
Re:Vector databases: analyzing the trade-offs
by tomhamer on Sunday August 20, 2023 @11:07PM
(#37205045)
Just to quickly add to ukuina's comment, marqo.ai does embedding generation and vector search end to end, so you can put in documents and the embeddings are automatically generated.
Re:Vector databases: analyzing the trade-offs
by minimaxir on Sunday August 20, 2023 @11:40PM
(#37205203)
SentenceTransformers all-MiniLM-L6-v2 is still your best bet since you can generate them in batches with GPU acceleration.
Re:Vector databases: analyzing the trade-offs
by dmezzetti on Monday August 21, 2023 @06:32AM
(#37207573)
Lot of tools that can do this and they've long been around. For example, txtai has been able to generate embeddings with sentence-transformers since 2020.
Re:Vector databases: analyzing the trade-offs
by kordlessagain on Monday August 21, 2023 @08:46AM
(#37208781)
I work at FeatureBase and I'm storing vectors from the Instructor Large library/model into our solution. Getting good results, which I should probably quantify at some point. One thing that FeatureBase does well is allow filtering of the vector space via SQL.
I would say that most people seem to prefer an engine that embeds and stores things as a service, but using Instructor is only a few lines of code and runs locally.