liuyu/pgvector
1
0
Fork 0
mirror of https://github.com/pgvector/pgvector.git synced 2026-07-11 23:26:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Improved HNSW recall tests - #535

Browse Source
This commit is contained in:
Andrew Kane
2024-04-26 13:08:48 -07:00
parent 78d32943ac
commit 48e68e5e42
6 changed files with 16 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
test/t/012_hnsw_vector_build_recall.pl
View File

@@ -8,6 +8,7 @@ my $node;
my @queries = (); my @queries = ();
my @expected; my @expected;
my $limit = 20; my $limit = 20;
my $array_sql = join(",", ('random() * random()') x 3);
sub test_recall sub test_recall
{ {
@@ -54,7 +55,7 @@ $node->start;
$node->safe_psql("postgres", "CREATE EXTENSION vector;"); $node->safe_psql("postgres", "CREATE EXTENSION vector;");
$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector(3));"); $node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector(3));");
$node->safe_psql("postgres", $node->safe_psql("postgres",
"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 10000) i;" "INSERT INTO tst SELECT i, ARRAY[$array_sql] FROM generate_series(1, 10000) i;"
); );
# Generate queries # Generate queries
@@ -90,7 +91,7 @@ for my $i (0 .. $#operators)
)); ));
# Test approximate results # Test approximate results
my $min = $operator eq "<#>" ? 0.80 : 0.99; my $min = 0.99;
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");

5
test/t/013_hnsw_vector_insert_recall.pl
View File

@@ -8,6 +8,7 @@ my $node;
my @queries = (); my @queries = ();
my @expected; my @expected;
my $limit = 20; my $limit = 20;
my $array_sql = join(",", ('random() * random()') x 3);
sub test_recall sub test_recall
{ {
@@ -83,7 +84,7 @@ for my $i (0 .. $#operators)
[qr{^$}], [qr{^$}],
"concurrent INSERTs", "concurrent INSERTs",
{ {
"013_hnsw_insert_recall_$opclass" => "INSERT INTO tst (v) VALUES (ARRAY[random(), random(), random()]);" "013_hnsw_insert_recall_$opclass" => "INSERT INTO tst (v) VALUES (ARRAY[$array_sql]);"
} }
); );
@@ -99,7 +100,7 @@ for my $i (0 .. $#operators)
} }
# Test approximate results # Test approximate results
my $min = $operator eq "<#>" ? 0.80 : 0.99; my $min = 0.99;
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");

4
test/t/024_hnsw_halfvec_build_recall.pl
View File

@@ -9,7 +9,7 @@ my @queries = ();
my @expected; my @expected;
my $limit = 20; my $limit = 20;
my $dim = 10; my $dim = 10;
my $array_sql = join(",", ('random()') x $dim); my $array_sql = join(",", ('2 * random() * random()') x $dim);
sub test_recall sub test_recall
{ {
@@ -94,7 +94,7 @@ for my $i (0 .. $#operators)
)); ));
# Test approximate results # Test approximate results
my $min = $operator eq "<#>" ? 0.93 : 0.98; my $min = 0.98;
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");

4
test/t/025_hnsw_halfvec_insert_recall.pl
View File

@@ -9,7 +9,7 @@ my @queries = ();
my @expected; my @expected;
my $limit = 20; my $limit = 20;
my $dim = 10; my $dim = 10;
my $array_sql = join(",", ('random()') x $dim); my $array_sql = join(",", ('2 * random() * random()') x $dim);
sub test_recall sub test_recall
{ {
@@ -103,7 +103,7 @@ for my $i (0 .. $#operators)
} }
# Test approximate results # Test approximate results
my $min = $operator eq "<#>" ? 0.94 : 0.98; my $min = 0.98;
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");

5
test/t/028_hnsw_sparsevec_build_recall.pl
View File

@@ -8,6 +8,7 @@ my $node;
my @queries = (); my @queries = ();
my @expected; my @expected;
my $limit = 20; my $limit = 20;
my $array_sql = join(",", ('random() * random()') x 3);
sub test_recall sub test_recall
{ {
@@ -54,7 +55,7 @@ $node->start;
$node->safe_psql("postgres", "CREATE EXTENSION vector;"); $node->safe_psql("postgres", "CREATE EXTENSION vector;");
$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v sparsevec(3));"); $node->safe_psql("postgres", "CREATE TABLE tst (i int4, v sparsevec(3));");
$node->safe_psql("postgres", $node->safe_psql("postgres",
"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()]::vector::sparsevec FROM generate_series(1, 10000) i;" "INSERT INTO tst SELECT i, ARRAY[$array_sql]::vector::sparsevec FROM generate_series(1, 10000) i;"
); );
# Generate queries # Generate queries
@@ -90,7 +91,7 @@ for my $i (0 .. $#operators)
)); ));
# Test approximate results # Test approximate results
my $min = $operator eq "<#>" ? 0.80 : 0.99; my $min = 0.99;
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");

5
test/t/029_hnsw_sparsevec_insert_recall.pl
View File

@@ -8,6 +8,7 @@ my $node;
my @queries = (); my @queries = ();
my @expected; my @expected;
my $limit = 20; my $limit = 20;
my $array_sql = join(",", ('random() * random()') x 3);
sub test_recall sub test_recall
{ {
@@ -83,7 +84,7 @@ for my $i (0 .. $#operators)
[qr{^$}], [qr{^$}],
"concurrent INSERTs", "concurrent INSERTs",
{ {
"025_hnsw_sparsevec_insert_recall_$opclass" => "INSERT INTO tst (v) VALUES (ARRAY[random(), random(), random()]::vector::sparsevec);" "025_hnsw_sparsevec_insert_recall_$opclass" => "INSERT INTO tst (v) VALUES (ARRAY[$array_sql]::vector::sparsevec);"
} }
); );
@@ -99,7 +100,7 @@ for my $i (0 .. $#operators)
} }
# Test approximate results # Test approximate results
my $min = $operator eq "<#>" ? 0.79 : 0.99; my $min = 0.99;
test_recall($min, $operator); test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;"); $node->safe_psql("postgres", "DROP INDEX idx;");