From e9d410988485331741a0ecc2f990d51ef732b06b Mon Sep 17 00:00:00 2001 From: Stefan Webb Date: Fri, 6 Dec 2024 15:34:17 -0800 Subject: [PATCH 1/2] Update sparse_vector.md I believe you have to use the `BM_25` metric for full-text search, correct? Then we should update the note to say that you can only use `IP` or `BM_25` with sparse vectors --- site/en/userGuide/schema/sparse_vector.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/site/en/userGuide/schema/sparse_vector.md b/site/en/userGuide/schema/sparse_vector.md index 746c0eb69..1659c01c8 100644 --- a/site/en/userGuide/schema/sparse_vector.md +++ b/site/en/userGuide/schema/sparse_vector.md @@ -563,7 +563,7 @@ For more information on similarity search parameters, refer to [​Basic ANN Sea When using sparse vectors in Milvus, consider the following limits: -- Currently, only the __IP__ distance metric is supported for sparse vectors. The high dimensionality of sparse vectors makes L2 and cosine distance impractical. +- Currently, only the __IP__ and __BM_25__ distance metrics are supported for sparse vectors. The high dimensionality of sparse vectors makes L2 and cosine distance impractical. - For sparse vector fields, only the __SPARSE_INVERTED_INDEX__ and __SPARSE_WAND__ index types are supported. From 8018e39e8394105f6ad64f858c70c5d09b159e88 Mon Sep 17 00:00:00 2001 From: Stefan Webb Date: Fri, 6 Dec 2024 15:40:01 -0800 Subject: [PATCH 2/2] Update sparse_vector.md --- site/en/userGuide/schema/sparse_vector.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/site/en/userGuide/schema/sparse_vector.md b/site/en/userGuide/schema/sparse_vector.md index 1659c01c8..511dbc3b5 100644 --- a/site/en/userGuide/schema/sparse_vector.md +++ b/site/en/userGuide/schema/sparse_vector.md @@ -563,7 +563,7 @@ For more information on similarity search parameters, refer to [​Basic ANN Sea When using sparse vectors in Milvus, consider the following limits: -- Currently, only the __IP__ and __BM_25__ distance metrics are supported for sparse vectors. The high dimensionality of sparse vectors makes L2 and cosine distance impractical. +- Currently, only the __IP__ and __BM25__ (for full-text search) distance metrics are supported for sparse vectors. The high dimensionality of sparse vectors makes L2 and cosine distance impractical. - For sparse vector fields, only the __SPARSE_INVERTED_INDEX__ and __SPARSE_WAND__ index types are supported.