questions about sparse index #34806
-
Hi, I'd like use milvus 2.4.6 to perform a hybrid search using BGE-M3 model, I want to use both sparse vectors and dense vectors, my questions are focus on sparse vectors index type. |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 1 reply
-
SPARSE_WAND is usually better at speed. |
Beta Was this translation helpful? Give feedback.
-
Hi @Gladiator566, thanks for the question!
The choice between
The default value for Choice of I would suggest to start from |
Beta Was this translation helpful? Give feedback.
Hi @Gladiator566, thanks for the question!
SPARSE_INVERTED_INDEX
is the plain old inverted index: for each dimension, maintain a list of doc ids that is not zero in this dimension; at search time, only compute scores of docs in merged lists of the query's non zero dimensions.SPARSE_WAND
is similar toSPARSE_INVERTED_INDEX
in that they both have the same inverted index, butSPARSE_WAND
maintains more metadata in the index to allow the utilization of the WAND algorithm to greatly reduce the number of docs that needs to be evaluated. If you are interested in the detail, you may refer to the original paper or this slide.The ch…