liuyu/pgvector
1
0
Fork 0
mirror of https://github.com/pgvector/pgvector.git synced 2026-07-10 22:56:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Improved precision for cosine distance [skip ci]

Browse Source
This commit is contained in:
Andrew Kane
2024-09-23 18:50:10 -07:00
parent 0a2803130f
commit 22efa8ec22
4 changed files with 6 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
sql/vector.sql
View File

@@ -932,7 +932,7 @@ CREATE OPERATOR CLASS minivec_ip_ops
CREATE OPERATOR CLASS minivec_cosine_ops CREATE OPERATOR CLASS minivec_cosine_ops
FOR TYPE minivec USING hnsw AS FOR TYPE minivec USING hnsw AS
OPERATOR 1 <=> (minivec, minivec) FOR ORDER BY float_ops, OPERATOR 1 <=> (minivec, minivec) FOR ORDER BY float_ops,
FUNCTION 1 minivec_negative_inner_product(minivec, minivec), FUNCTION 1 cosine_distance(minivec, minivec),
FUNCTION 2 l2_norm(minivec), FUNCTION 2 l2_norm(minivec),
FUNCTION 3 hnsw_minivec_support(internal); FUNCTION 3 hnsw_minivec_support(internal);

6
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);
} }
@@ -1382,7 +1385,8 @@ hnsw_minivec_support(PG_FUNCTION_ARGS)
{ {
static const HnswTypeInfo typeInfo = { static const HnswTypeInfo typeInfo = {
.maxDimensions = HNSW_MAX_DIM * 4, .maxDimensions = HNSW_MAX_DIM * 4,
.normalize = minivec_l2_normalize, /* Do not normalize to maximize precision */
.normalize = NULL,
.checkValue = NULL .checkValue = NULL
}; };

4
test/t/039_hnsw_minivec_build_recall.pl
View File

@@ -95,10 +95,6 @@ for my $i (0 .. $#operators)
# Test approximate results # Test approximate results
my $min = 0.98; my $min = 0.98;
if ($operator eq "<=>")
{
$min = 0.60;
}
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");

4
test/t/040_hnsw_minivec_insert_recall.pl
View File

@@ -104,10 +104,6 @@ for my $i (0 .. $#operators)
# Test approximate results # Test approximate results
my $min = 0.98; my $min = 0.98;
if ($operator eq "<=>")
{
$min = 0.60;
}
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");