Qdrant
Custom metric for sparse vectors #5678
Unanswered
KukumavMozolo
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi there,
i implemented a custom weighted euclidean distance for sparse vectors.
Unfortunately when scoring vectors qdrant reports incorrect distances.
Could it be that using metrics where the distance/similarity does not only depend on shared indices breaks assumptions in HNSW?
Is there a fix?
Beta Was this translation helpful? Give feedback.
All reactions