Hi,
I tried using bulk search in my approximate kNN (aNN) searches to improve throughput and it did help.
However I cannot understand from the results which hit referes to which query_vector.
For example if I search for 2 query_vectors in the same search request, and specify k=5, I will get 10 hits.
I expected it to be in the format of "5 hits refer to query_vector1, 5 hits refer to query_vector2", But it is not.
Is there a way to achieve what I described? Or any plan to?
Thanks!