Hi all, I’m working with vector queries using KnnVectorQuery, which, as I understand it, requires a vector field with the target vector for vector search and a separate filter query for other fields based filters. This setup has worked well for me so far.
However, I have two queries or use-cases: 1. A use case where advanced filters are dynamically formed, combining both Lucene regular fields and vector fields (e.g., Content Field). PFA example. In this case, separating the vector field from the nested filter pattern doesn't seem to align well with the expected structure of KnnVectorQuery. 2. Second, Can Vector Field be used inside as a Filter Criteria currently ? Based on how vector search (approximate) works, does it make sense to use it? Has anyone encountered a similar challenge or have suggestions on how to approach these? Regards, Arun
--------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org
