Re:Storing OpenAI embeddings in Postgres with pgvector
by e12e on Wednesday February 08, 2023 @03:04AM
(#34705426)
> And the point of converting text or other objects (like images) into embeddings is to compare a large number of documents to a source document very fast.
But a "source document" could also be a natural language query that you need to convert to an embedding - if you want to enable natural language search on your corpus? (maybe along with handling queries in French getting good semantic hits in English etc?)?
But a "source document" could also be a natural language query that you need to convert to an embedding - if you want to enable natural language search on your corpus? (maybe along with handling queries in French getting good semantic hits in English etc?)?