adding tests

Author: benwtrent

server/src/main/java/org/elasticsearch/index/codec/vectors/DefaultIVFVectorsReader.java

@@ -20,8 +20,6 @@
 import org.apache.lucene.util.VectorUtil;
 import org.apache.lucene.util.hnsw.NeighborQueue;
 import org.elasticsearch.index.codec.vectors.reflect.OffHeapStats;
-import org.elasticsearch.logging.LogManager;
-import org.elasticsearch.logging.Logger;
 import org.elasticsearch.simdvec.ES91OSQVectorsScorer;
 import org.elasticsearch.simdvec.ES92Int7VectorsScorer;
 import org.elasticsearch.simdvec.ESVectorUtil;
@@ -42,8 +40,6 @@
  */
 public class DefaultIVFVectorsReader extends IVFVectorsReader implements OffHeapStats {
 
-    static Logger logger = LogManager.getLogger(DefaultIVFVectorsReader.class);
-
     // The percentage of centroids that are scored to keep recall
     public static final double CENTROID_SAMPLING_PERCENTAGE = 0.2;
 

server/src/main/java/org/elasticsearch/search/vectors/DiversifiedIVFKnnCollectorManager.java

@@ -11,9 +11,7 @@
 
 import org.apache.lucene.index.LeafReaderContext;
 import org.apache.lucene.search.IndexSearcher;
-import org.apache.lucene.search.KnnCollector;
 import org.apache.lucene.search.join.BitSetProducer;
-import org.apache.lucene.search.knn.KnnCollectorManager;
 import org.apache.lucene.search.knn.KnnSearchStrategy;
 import org.apache.lucene.util.BitSet;
 
@@ -30,7 +28,8 @@ public class DiversifiedIVFKnnCollectorManager extends AbstractIVFKnnVectorQuery
     }
 
     @Override
-    public AbstractMaxScoreKnnCollector newCollector(int visitedLimit, KnnSearchStrategy searchStrategy, LeafReaderContext context) throws IOException {
+    public AbstractMaxScoreKnnCollector newCollector(int visitedLimit, KnnSearchStrategy searchStrategy, LeafReaderContext context)
+        throws IOException {
         BitSet parentBitSet = parentsFilter.getBitSet(context);
         if (parentBitSet == null) {
             return null;

server/src/main/java/org/elasticsearch/search/vectors/DiversifyingNearestChildrenKnnCollector.java

@@ -105,7 +105,9 @@ public int numCollected() {
 
     @Override
     public long getMinCompetitiveDocScore() {
-        return heap.size() > 0 ? Math.max(NeighborQueue.encodeRaw(heap.topNode(), heap.topScore()), minCompetitiveDocScore) : minCompetitiveDocScore;
+        return heap.size() > 0
+            ? Math.max(NeighborQueue.encodeRaw(heap.topNode(), heap.topScore()), minCompetitiveDocScore)
+            : minCompetitiveDocScore;
     }
 
     @Override

server/src/main/java/org/elasticsearch/search/vectors/MaxScoreTopKnnCollector.java

@@ -15,13 +15,13 @@
 import org.apache.lucene.search.knn.KnnSearchStrategy;
 import org.elasticsearch.index.codec.vectors.cluster.NeighborQueue;
 
-public class MaxScoreTopKnnCollector extends AbstractMaxScoreKnnCollector {
+class MaxScoreTopKnnCollector extends AbstractMaxScoreKnnCollector {
 
     private long minCompetitiveDocScore;
     private float minCompetitiveSimilarity;
     protected final NeighborQueue queue;
 
-    public MaxScoreTopKnnCollector(int k, long visitLimit, KnnSearchStrategy searchStrategy) {
+    MaxScoreTopKnnCollector(int k, long visitLimit, KnnSearchStrategy searchStrategy) {
         super(k, visitLimit, searchStrategy);
         this.minCompetitiveDocScore = LEAST_COMPETITIVE;
         this.minCompetitiveSimilarity = Float.NEGATIVE_INFINITY;

server/src/test/java/org/elasticsearch/search/vectors/IVFKnnSearchStrategyTests.java

@@ -0,0 +1,41 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the "Elastic License
+ * 2.0", the "GNU Affero General Public License v3.0 only", and the "Server Side
+ * Public License v 1"; you may not use this file except in compliance with, at
+ * your election, the "Elastic License 2.0", the "GNU Affero General Public
+ * License v3.0 only", or the "Server Side Public License, v 1".
+ */
+
+package org.elasticsearch.search.vectors;
+
+import org.elasticsearch.index.codec.vectors.cluster.NeighborQueue;
+import org.elasticsearch.test.ESTestCase;
+
+import java.util.concurrent.atomic.LongAccumulator;
+
+public class IVFKnnSearchStrategyTests extends ESTestCase {
+
+    public void testMaxScorePropagation() {
+        LongAccumulator accumulator = new LongAccumulator(Long::max, AbstractMaxScoreKnnCollector.LEAST_COMPETITIVE);
+        IVFKnnSearchStrategy strategy = new IVFKnnSearchStrategy(0.5f, accumulator);
+        MaxScoreTopKnnCollector collector = new MaxScoreTopKnnCollector(2, 1000, strategy);
+        strategy.setCollector(collector);
+
+        collector.collect(1, 0.9f);
+        long competitiveScore = NeighborQueue.encodeRaw(1, 0.9f);
+
+        // accumulator should now be updated
+        strategy.nextVectorsBlock();
+        assertEquals(competitiveScore, accumulator.get());
+        assertEquals(competitiveScore, collector.getMinCompetitiveDocScore());
+
+        // updated accumulator directly with more competitive score
+        competitiveScore = NeighborQueue.encodeRaw(2, 1.5f);
+        accumulator.accumulate(competitiveScore);
+        assertEquals(competitiveScore, accumulator.get());
+        strategy.nextVectorsBlock();
+        assertEquals(competitiveScore, collector.getMinCompetitiveDocScore());
+        assertEquals(competitiveScore, accumulator.get());
+    }
+}

server/src/test/java/org/elasticsearch/search/vectors/MaxScoreTopKnnCollectorTests.java

@@ -0,0 +1,47 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the "Elastic License
+ * 2.0", the "GNU Affero General Public License v3.0 only", and the "Server Side
+ * Public License v 1"; you may not use this file except in compliance with, at
+ * your election, the "Elastic License 2.0", the "GNU Affero General Public
+ * License v3.0 only", or the "Server Side Public License, v 1".
+ */
+
+package org.elasticsearch.search.vectors;
+
+import org.elasticsearch.index.codec.vectors.cluster.NeighborQueue;
+import org.elasticsearch.test.ESTestCase;
+
+import java.util.concurrent.atomic.LongAccumulator;
+
+public class MaxScoreTopKnnCollectorTests extends ESTestCase {
+
+    public void testMaxScorePropagation() {
+        LongAccumulator accumulator = new LongAccumulator(Long::max, AbstractMaxScoreKnnCollector.LEAST_COMPETITIVE);
+        MaxScoreTopKnnCollector collector = new MaxScoreTopKnnCollector(2, 1000, new IVFKnnSearchStrategy(0.5f, accumulator));
+        long competitiveScore = NeighborQueue.encodeRaw(1, 0.9f);
+
+        collector.updateMinCompetitiveDocScore(competitiveScore);
+        assertEquals(competitiveScore, collector.getMinCompetitiveDocScore());
+        // haven't collected k results
+        assertTrue(Float.NEGATIVE_INFINITY == collector.minCompetitiveSimilarity());
+        collector.collect(2, 1.5f);
+        assertTrue(Float.NEGATIVE_INFINITY == collector.minCompetitiveSimilarity());
+
+        // we always provide the min competitive that this collector collected
+        assertEquals(NeighborQueue.encodeRaw(2, 1.5f), collector.getMinCompetitiveDocScore());
+        collector.collect(3, 1.9f);
+
+        // min competitive for this collector is more than global
+        assertEquals(NeighborQueue.encodeRaw(2, 1.5f), collector.getMinCompetitiveDocScore());
+
+        // we have collected k results, min competitive is the min value collected
+        assertEquals(1.5f, collector.minCompetitiveSimilarity(), 0.0f);
+        // we update the global min competitive doc score with a new value
+        competitiveScore = NeighborQueue.encodeRaw(4, 4f);
+        collector.updateMinCompetitiveDocScore(competitiveScore);
+        assertEquals(competitiveScore, collector.getMinCompetitiveDocScore());
+        assertEquals(4f, collector.minCompetitiveSimilarity(), 0.0f);
+    }
+
+}