Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fixes #974 - returning score driven by cosine similarity by (1 - distance) instead of distance #1048

Merged
merged 1 commit into from
Oct 24, 2024
Merged

Fixes #974 - returning score driven by cosine similarity by (1 - distance) instead of distance #1048

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion vectorstores/pgvector/pgvector.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -318,7 +318,7 @@ func (s Store) SimilaritySearch(
SELECT
data.document,
data.cmetadata,
data.distance
(1 - data.distance) AS score
FROM (
SELECT
filtered_embedding_dims.*,
Expand Down
45 changes: 45 additions & 0 deletions vectorstores/pgvector/pgvector_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -179,6 +179,51 @@ func TestPgvectorStoreRestWithScoreThreshold(t *testing.T) {
require.Len(t, docs, 10)
}

func TestPgvectorStoreSimilarityScore(t *testing.T) {
t.Parallel()
pgvectorURL := preCheckEnvSetting(t)
ctx := context.Background()

llm, err := openai.New(
openai.WithEmbeddingModel("text-embedding-ada-002"),
)
require.NoError(t, err)
e, err := embeddings.NewEmbedder(llm)
require.NoError(t, err)

conn, err := pgx.Connect(ctx, pgvectorURL)
require.NoError(t, err)

store, err := pgvector.New(
ctx,
pgvector.WithConn(conn),
pgvector.WithEmbedder(e),
pgvector.WithPreDeleteCollection(true),
pgvector.WithCollectionName(makeNewCollectionName()),
)
require.NoError(t, err)

defer cleanupTestArtifacts(ctx, t, store, pgvectorURL)

_, err = store.AddDocuments(context.Background(), []schema.Document{
{PageContent: "Tokyo is the capital city of Japan."},
{PageContent: "Paris is the city of love."},
{PageContent: "I like to visit London."},
})
require.NoError(t, err)

// test with a score threshold of 0.8, expected 6 documents
docs, err := store.SimilaritySearch(
ctx,
"What is the capital city of Japan?",
3,
vectorstores.WithScoreThreshold(0.8),
)
require.NoError(t, err)
require.Len(t, docs, 1)
require.True(t, docs[0].Score > 0.9)
}

func TestSimilaritySearchWithInvalidScoreThreshold(t *testing.T) {
t.Parallel()
pgvectorURL := preCheckEnvSetting(t)
Expand Down