liuyu/pgvector
1
0
Fork 0
mirror of https://github.com/pgvector/pgvector.git synced 2026-07-10 06:40:57 +08:00
Code Issues Packages Projects Releases Wiki Activity

Exclude zero vectors for cosine distance to be consistent with other types [skip ci]

Browse Source
This commit is contained in:
Andrew Kane
2024-09-23 13:35:36 -07:00
parent 0296a08ccf
commit ffca8e720c
5 changed files with 16 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
README.md
View File

@@ -953,8 +953,9 @@ Function | Description | Added
--- | --- | --- --- | --- | ---
cosine_distance(intvec, intvec) → double precision | cosine distance | 0.8.0 cosine_distance(intvec, intvec) → double precision | cosine distance | 0.8.0
inner_product(intvec, intvec) → double precision | inner product | 0.8.0 inner_product(intvec, intvec) → double precision | inner product | 0.8.0
l2_distance(intvec, intvec) → double precision | Euclidean distance | 0.8.0
l1_distance(intvec, intvec) → double precision | taxicab distance | 0.8.0 l1_distance(intvec, intvec) → double precision | taxicab distance | 0.8.0
l2_distance(intvec, intvec) → double precision | Euclidean distance | 0.8.0
l2_norm(intvec) → double precision | Euclidean norm | 0.8.0
### Bit Type ### Bit Type

4
sql/vector--0.7.4--0.8.0.sql
View File

@@ -42,6 +42,9 @@ CREATE FUNCTION cosine_distance(intvec, intvec) RETURNS float8
CREATE FUNCTION l1_distance(intvec, intvec) RETURNS float8 CREATE FUNCTION l1_distance(intvec, intvec) RETURNS float8
AS 'MODULE_PATHNAME', 'intvec_l1_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE; AS 'MODULE_PATHNAME', 'intvec_l1_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
CREATE FUNCTION l2_norm(intvec) RETURNS float8
AS 'MODULE_PATHNAME', 'intvec_l2_norm' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
CREATE FUNCTION intvec_l2_squared_distance(intvec, intvec) RETURNS float8 CREATE FUNCTION intvec_l2_squared_distance(intvec, intvec) RETURNS float8
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE; AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
@@ -96,6 +99,7 @@ CREATE OPERATOR CLASS intvec_cosine_ops
FOR TYPE intvec USING hnsw AS FOR TYPE intvec USING hnsw AS
OPERATOR 1 <=> (intvec, intvec) FOR ORDER BY float_ops, OPERATOR 1 <=> (intvec, intvec) FOR ORDER BY float_ops,
FUNCTION 1 cosine_distance(intvec, intvec), FUNCTION 1 cosine_distance(intvec, intvec),
FUNCTION 2 l2_norm(intvec),
FUNCTION 3 hnsw_intvec_support(internal); FUNCTION 3 hnsw_intvec_support(internal);
CREATE OPERATOR CLASS intvec_l1_ops CREATE OPERATOR CLASS intvec_l1_ops

4
sql/vector.sql
View File

@@ -692,6 +692,9 @@ CREATE FUNCTION cosine_distance(intvec, intvec) RETURNS float8
CREATE FUNCTION l1_distance(intvec, intvec) RETURNS float8 CREATE FUNCTION l1_distance(intvec, intvec) RETURNS float8
AS 'MODULE_PATHNAME', 'intvec_l1_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE; AS 'MODULE_PATHNAME', 'intvec_l1_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
CREATE FUNCTION l2_norm(intvec) RETURNS float8
AS 'MODULE_PATHNAME', 'intvec_l2_norm' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
-- intvec private functions -- intvec private functions
CREATE FUNCTION intvec_l2_squared_distance(intvec, intvec) RETURNS float8 CREATE FUNCTION intvec_l2_squared_distance(intvec, intvec) RETURNS float8
@@ -756,6 +759,7 @@ CREATE OPERATOR CLASS intvec_cosine_ops
FOR TYPE intvec USING hnsw AS FOR TYPE intvec USING hnsw AS
OPERATOR 1 <=> (intvec, intvec) FOR ORDER BY float_ops, OPERATOR 1 <=> (intvec, intvec) FOR ORDER BY float_ops,
FUNCTION 1 cosine_distance(intvec, intvec), FUNCTION 1 cosine_distance(intvec, intvec),
FUNCTION 2 l2_norm(intvec),
FUNCTION 3 hnsw_intvec_support(internal); FUNCTION 3 hnsw_intvec_support(internal);
CREATE OPERATOR CLASS intvec_l1_ops CREATE OPERATOR CLASS intvec_l1_ops

3
src/hnswutils.c
View File

@@ -159,6 +159,9 @@ HnswOptionalProcInfo(Relation index, uint16 procnum)
Datum Datum
HnswNormValue(const HnswTypeInfo * typeInfo, Oid collation, Datum value) HnswNormValue(const HnswTypeInfo * typeInfo, Oid collation, Datum value)
{ {
if (!typeInfo->normalize)
return value;
return DirectFunctionCall1Coll(typeInfo->normalize, collation, value); return DirectFunctionCall1Coll(typeInfo->normalize, collation, value);
} }

7
test/expected/hnsw_intvec.out
View File

@@ -64,19 +64,18 @@ SELECT * FROM t ORDER BY val <=> '[3,3,3]';
[1,1,1] [1,1,1]
[1,2,3] [1,2,3]
[1,2,4] [1,2,4]
[0,0,0] (3 rows)
(4 rows)
SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> '[0,0,0]') t2; SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> '[0,0,0]') t2;
count count
------- -------
4 3
(1 row) (1 row)
SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::intvec)) t2; SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::intvec)) t2;
count count
------- -------
4 3
(1 row) (1 row)
DROP TABLE t; DROP TABLE t;