You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
#33269
Create a vector for the count field, for example [count, 0]. Then search for the farthest verters and sort them based on similarity, but distance it is already overflowed。
Describe the solution you'd like.
example1:
search(vectors_to_search, "embeddings", search_params, limit=3000, output_fields=["embeddings","count"],expr="order by count")
example2:
select embeddings, count from table order by count
Describe an alternate solution.
No response
Anything else? (Additional Context)
No response
The text was updated successfully, but these errors were encountered:
how why do need to put count in a vector field? does't seems to make sense?
为什么要在向量场中加入计数呢? 似乎说不通?
search vector, Then order by count, but it seems that milvus does not have the ability to sort scalars.
So,create a count in a vector field and sort it by similarity.
Is there an existing issue for this?
Is your feature request related to a problem? Please describe.
#33269
Create a vector for the count field, for example [count, 0]. Then search for the farthest verters and sort them based on similarity, but distance it is already overflowed。
Describe the solution you'd like.
example1:
search(vectors_to_search, "embeddings", search_params, limit=3000, output_fields=["embeddings","count"],expr="order by count")
example2:
select embeddings, count from table order by count
Describe an alternate solution.
No response
Anything else? (Additional Context)
No response
The text was updated successfully, but these errors were encountered: