From 6597eb1ea614fbd8345eccd475de39060055dd2e Mon Sep 17 00:00:00 2001 From: Vijayan Balasubramanian Date: Thu, 26 Dec 2024 11:13:20 -0800 Subject: [PATCH] Have one score definition for cosinesimilarity Currently we have different score calculation for cosine similarity, for ex: script score, approximate search, exact search has diffent formula to convert distance to cosine similarity that is aligned with OpenSearch score. To keep it consistent, we will be using one defintion which is used by Lucene as standard definition for cosine similarity for all search types. Signed-off-by: Vijayan Balasubramanian --- CHANGELOG.md | 1 + .../org/opensearch/knn/index/SpaceType.java | 14 ++++- .../knn/plugin/script/KNNScoringSpace.java | 7 ++- .../org/opensearch/knn/index/NmslibIT.java | 58 +++++++++++++++++++ .../plugin/script/KNNScoringSpaceTests.java | 7 ++- 5 files changed, 84 insertions(+), 3 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index a19b53fd8..855c239a4 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -23,6 +23,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), - Introduced a writing layer in native engines where relies on the writing interface to process IO. (#2241)[https://github.com/opensearch-project/k-NN/pull/2241] - Allow method parameter override for training based indices (#2290) https://github.com/opensearch-project/k-NN/pull/2290] - Optimizes lucene query execution to prevent unnecessary rewrites (#2305)[https://github.com/opensearch-project/k-NN/pull/2305] +- Use one formula to calculate cosine similarity (#2357)[https://github.com/opensearch-project/k-NN/pull/2357] ### Bug Fixes * Fixing the bug when a segment has no vector field present for disk based vector search (#2282)[https://github.com/opensearch-project/k-NN/pull/2282] * Allow validation for non knn index only after 2.17.0 (#2315)[https://github.com/opensearch-project/k-NN/pull/2315] diff --git a/src/main/java/org/opensearch/knn/index/SpaceType.java b/src/main/java/org/opensearch/knn/index/SpaceType.java index abe265a01..5d90071e8 100644 --- a/src/main/java/org/opensearch/knn/index/SpaceType.java +++ b/src/main/java/org/opensearch/knn/index/SpaceType.java @@ -60,9 +60,21 @@ public float scoreToDistanceTranslation(float score) { } }, COSINESIMIL("cosinesimil") { + /** + * Cosine similarity has range of [-1, 1] where -1 represents vectors are at diametrically opposite, and 1 is where + * they are identical in direction and perfectly similar. In Lucene, scores have to be in the range of [0, Float.MAX_VALUE]. + * Hence, to move the range from [-1, 1] to [ 0, Float.MAX_VALUE], we convert using following formula which is adopted + * by Lucene as mentioned here + * https://github.com/apache/lucene/blob/0494c824e0ac8049b757582f60d085932a890800/lucene/core/src/java/org/apache/lucene/index/VectorSimilarityFunction.java#L73 + * We expect raw score = 1 - cosine(x,y), if underlying library returns different range or other than expected raw score, + * they should override this method to either provide valid range or convert raw score to the format as 1 - cosine and call this method + * + * @param rawScore score returned from underlying library + * @return Lucene scaled score + */ @Override public float scoreTranslation(float rawScore) { - return 1 / (1 + rawScore); + return Math.max((2.0F - rawScore) / 2.0F, 0.0F); } @Override diff --git a/src/main/java/org/opensearch/knn/plugin/script/KNNScoringSpace.java b/src/main/java/org/opensearch/knn/plugin/script/KNNScoringSpace.java index 71616c9fd..9744796c6 100644 --- a/src/main/java/org/opensearch/knn/plugin/script/KNNScoringSpace.java +++ b/src/main/java/org/opensearch/knn/plugin/script/KNNScoringSpace.java @@ -144,7 +144,12 @@ public CosineSimilarity(Object query, MappedFieldType fieldType) { protected BiFunction getScoringMethod(final float[] processedQuery) { SpaceType.COSINESIMIL.validateVector(processedQuery); float qVectorSquaredMagnitude = getVectorMagnitudeSquared(processedQuery); - return (float[] q, float[] v) -> 1 + KNNScoringUtil.cosinesimilOptimized(q, v, qVectorSquaredMagnitude); + // To be consistent, we will be using same formula used by lucene as mentioned below + // https://github.com/apache/lucene/blob/0494c824e0ac8049b757582f60d085932a890800/lucene/core/src/java/org/apache/lucene/index/VectorSimilarityFunction.java#L73 + return (float[] q, float[] v) -> Math.max( + (1.0F + KNNScoringUtil.cosinesimilOptimized(q, v, qVectorSquaredMagnitude)) / 2.0F, + 0.0F + ); } } diff --git a/src/test/java/org/opensearch/knn/index/NmslibIT.java b/src/test/java/org/opensearch/knn/index/NmslibIT.java index 8ca436bf4..e2e7613a2 100644 --- a/src/test/java/org/opensearch/knn/index/NmslibIT.java +++ b/src/test/java/org/opensearch/knn/index/NmslibIT.java @@ -195,6 +195,64 @@ public void testEndToEnd() throws Exception { fail("Graphs are not getting evicted"); } + public void testEndToEnd_withApproxAndExactSearch_inSameIndex_ForCosineSpaceType() throws Exception { + String indexName = "test-index-1"; + String fieldName = "test-field-1"; + SpaceType spaceType = SpaceType.COSINESIMIL; + Integer dimension = testData.indexData.vectors[0].length; + + // Create an index + XContentBuilder builder = XContentFactory.jsonBuilder() + .startObject() + .startObject("properties") + .startObject(fieldName) + .field("type", "knn_vector") + .field("dimension", dimension) + .field(KNNConstants.METHOD_PARAMETER_SPACE_TYPE, spaceType.getValue()) + .startObject(KNNConstants.KNN_METHOD) + .field(KNNConstants.NAME, KNNConstants.METHOD_HNSW) + .field(KNNConstants.KNN_ENGINE, KNNEngine.NMSLIB.getName()) + .endObject() + .endObject() + .endObject() + .endObject(); + + Map mappingMap = xContentBuilderToMap(builder); + String mapping = builder.toString(); + + createKnnIndex(indexName, buildKNNIndexSettings(0), mapping); + + // Index one document + addKnnDoc(indexName, randomAlphaOfLength(5), fieldName, Floats.asList(testData.indexData.vectors[0]).toArray()); + + // Assert we have the right number of documents in the index + refreshAllIndices(); + assertEquals(1, getDocCount(indexName)); + // update threshold setting to skip building graph + updateIndexSettings(indexName, Settings.builder().put(KNNSettings.INDEX_KNN_ADVANCED_APPROXIMATE_THRESHOLD, -1)); + // add duplicate document with different id + addKnnDoc(indexName, randomAlphaOfLength(5), fieldName, Floats.asList(testData.indexData.vectors[0]).toArray()); + assertEquals(2, getDocCount(indexName)); + final int k = 2; + // search index + Response response = searchKNNIndex( + indexName, + KNNQueryBuilder.builder().fieldName(fieldName).vector(testData.queries[0]).k(k).build(), + k + ); + String responseBody = EntityUtils.toString(response.getEntity()); + List knnResults = parseSearchResponse(responseBody, fieldName); + assertEquals(k, knnResults.size()); + + List actualScores = parseSearchResponseScore(responseBody, fieldName); + + // both document should have identical score + assertEquals(actualScores.get(0), actualScores.get(1), 0.001); + + // Delete index + deleteKNNIndex(indexName); + } + @SneakyThrows private void validateSearch( final String indexName, diff --git a/src/test/java/org/opensearch/knn/plugin/script/KNNScoringSpaceTests.java b/src/test/java/org/opensearch/knn/plugin/script/KNNScoringSpaceTests.java index 4fc549d6b..99e847eea 100644 --- a/src/test/java/org/opensearch/knn/plugin/script/KNNScoringSpaceTests.java +++ b/src/test/java/org/opensearch/knn/plugin/script/KNNScoringSpaceTests.java @@ -10,6 +10,7 @@ import java.util.Locale; import lombok.SneakyThrows; +import org.apache.lucene.index.VectorSimilarityFunction; import org.opensearch.index.mapper.MappedFieldType; import org.opensearch.knn.KNNTestCase; import org.opensearch.knn.index.engine.KNNMethodContext; @@ -86,7 +87,11 @@ public void testCosineSimilarity_whenValid_thenSucceed() { getMappingConfigForMethodMapping(knnMethodContext, 3) ); KNNScoringSpace.CosineSimilarity cosineSimilarity = new KNNScoringSpace.CosineSimilarity(arrayListQueryObject, fieldType); - assertEquals(2F, cosineSimilarity.getScoringMethod().apply(arrayFloat2, arrayFloat), 0.1F); + assertEquals( + VectorSimilarityFunction.COSINE.compare(arrayFloat2, arrayFloat), + cosineSimilarity.getScoringMethod().apply(arrayFloat2, arrayFloat), + 0.1F + ); // invalid zero vector final List queryZeroVector = List.of(0.0f, 0.0f, 0.0f);