Misc cleanup of KMPPClusterer after recent change. Expose MAX_PQ_TRAI…

…NING_SET_SIZE.
jbellis · Feb 8, 2024 · e800853 · e800853
1 parent 335a24d
commit e800853
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 8 deletions.
diff --git a/jvector-base/src/main/java/io/github/jbellis/jvector/pq/KMeansPlusPlusClusterer.java b/jvector-base/src/main/java/io/github/jbellis/jvector/pq/KMeansPlusPlusClusterer.java
@@ -20,7 +20,6 @@
 
 import java.util.Arrays;
 import java.util.List;
-import java.util.Random;
 import java.util.concurrent.ThreadLocalRandom;
 
 /**
@@ -40,20 +39,22 @@ public class KMeansPlusPlusClusterer {
     private final float[][] centroidNums;
 
     /**
-     * Constructs a KMeansPlusPlusFloatClusterer with the specified number of clusters,
-     * maximum iterations, and distance function.
+     * Constructs a KMeansPlusPlusFloatClusterer with the specified points and number of clusters.
      *
+     * @param points the points to cluster.
      * @param k number of clusters.
      */
     public KMeansPlusPlusClusterer(float[][] points, int k) {
         this(points, chooseInitialCentroids(points, k));
     }
 
     /**
-     * Constructs a KMeansPlusPlusFloatClusterer with the specified number of clusters,
-     * maximum iterations, and distance function.
+     * Constructs a KMeansPlusPlusFloatClusterer with the specified points and initial centroids.
      * <p>
      * The initial centroids provided as a parameter are copied before modification.
+     *
+     * @param points the points to cluster.
+     * @param centroids the initial centroids.
      */
     public KMeansPlusPlusClusterer(float[][] points, float[][] centroids) {
         this.points = points;

diff --git a/jvector-base/src/main/java/io/github/jbellis/jvector/pq/ProductQuantization.java b/jvector-base/src/main/java/io/github/jbellis/jvector/pq/ProductQuantization.java
@@ -19,9 +19,7 @@
 import io.github.jbellis.jvector.disk.Io;
 import io.github.jbellis.jvector.disk.RandomAccessReader;
 import io.github.jbellis.jvector.graph.RandomAccessVectorValues;
-import io.github.jbellis.jvector.util.ExplicitThreadLocal;
 import io.github.jbellis.jvector.util.PhysicalCoreExecutor;
-import io.github.jbellis.jvector.util.ExplicitThreadLocal;
 import io.github.jbellis.jvector.util.RamUsageEstimator;
 import io.github.jbellis.jvector.vector.VectorUtil;
 
@@ -47,7 +45,7 @@
 public class ProductQuantization implements VectorCompressor<byte[]> {
     static final int CLUSTERS = 256; // number of clusters per subspace = one byte's worth
     static final int K_MEANS_ITERATIONS = 6;
-    static final int MAX_PQ_TRAINING_SET_SIZE = 128000;
+    public static final int MAX_PQ_TRAINING_SET_SIZE = 128000;
 
     final float[][][] codebooks;
     private final int M; // codebooks.length, redundantly reproduced for convenience