Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/5267#discussion_r27613304
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/clustering/HierarchicalClustering.scala
---
@@ -0,0 +1,610 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.mllib.clustering
+
+import breeze.linalg.{DenseVector => BDV, SparseVector => BSV, Vector =>
BV, norm => breezeNorm}
+import org.apache.spark.mllib.linalg.{Vector, Vectors}
+import org.apache.spark.rdd.RDD
+import org.apache.spark.util.random.XORShiftRandom
+import org.apache.spark.{Logging, SparkException}
+
+import scala.collection.{Map, mutable}
+
+
+/**
+ * Top-level methods for calling the hierarchical clustering algorithm
+ */
+object HierarchicalClustering extends Logging {
+
+ private[clustering] val ROOT_INDEX_KEY: Long = 1
+
+ /**
+ * Trains a hierarchical clustering model with the given data
+ *
+ * @param data trained data
+ * @param numClusters the maximum number of clusters you want
+ * @return a hierarchical clustering model
+ */
+ def train(data: RDD[Vector], numClusters: Int):
HierarchicalClusteringModel = {
+ val algo = new HierarchicalClustering().setNumClusters(numClusters)
+ algo.run(data)
+ }
+
+ /**
+ * Trains a hierarchical clustering model with the given data
+ *
+ * @param data training data
+ * @param numClusters the maximum number of clusters you want
+ * @param maxIterations the number of maximal iterations
+ * @param maxRetries the number of maximum retries when the clustering
can't be succeeded
+ * @param seed the randomseed to generate the initial vectors for each
bisecting
+ * @return a hierarchical clustering model
+ */
+ def train(data: RDD[Vector],
+ numClusters: Int,
+ maxIterations: Int,
+ maxRetries: Int,
+ seed: Int): HierarchicalClusteringModel = {
+
+ val algo = new HierarchicalClustering().setNumClusters(numClusters)
+ .setMaxIterations(maxIterations)
+ .setMaxRetries(maxRetries)
+ .setSeed(seed)
+ algo.run(data)
+ }
+
+ /**
+ * Finds the closes cluster's center
+ *
+ * @param metric a distance metric
+ * @param centers centers of the clusters
+ * @param point a target point
+ * @return an index of the array of clusters
+ */
+ private[mllib]
+ def findClosestCenter(metric: Function2[BV[Double], BV[Double], Double])
+ (centers: Seq[BV[Double]])(point: BV[Double]): Int = {
+ val (closestCenter, closestIndex) =
+ centers.zipWithIndex.map { case (center, idx) => (metric(center,
point), idx)}.minBy(_._1)
+ closestIndex
+ }
+}
+
+/**
+ * This is a divisive hierarchical clustering algorithm based on bi-sect
k-means algorithm.
+ *
+ * The main idea of this algorithm is based on "A comparison of document
clustering techniques",
+ * M. Steinbach, G. Karypis and V. Kumar. Workshop on Text Mining, KDD,
2000.
+ * http://cs.fit.edu/~pkc/classes/ml-internet/papers/steinbach00tr.pdf
+ *
+ * @param numClusters tne number of clusters you want
+ * @param clusterMap the pairs of cluster and its index as Map
+ * @param maxIterations the number of maximal iterations
+ * @param maxRetries the number of maximum retries
+ * @param seed a random seed
+ */
+class HierarchicalClustering(
+ private var numClusters: Int,
+ private var clusterMap: Map[Long, ClusterTree],
+ private var maxIterations: Int,
+ private var maxRetries: Int,
+ private var seed: Long) extends Logging {
+
+ /**
+ * Constructs with the default configuration
+ */
+ def this() = this(20, mutable.ListMap.empty[Long, ClusterTree], 20, 10,
1)
+
+ /**
+ * Sets the number of clusters you want
+ */
+ def setNumClusters(numClusters: Int): this.type = {
+ this.numClusters = numClusters
+ this
+ }
+
+ /**
+ * Sets the number of maximal iterations in each clustering step
+ */
+ def setMaxIterations(maxIterations: Int): this.type = {
+ this.maxIterations = maxIterations
+ this
+ }
+
+ def getSubIterations(): Int = this.maxIterations
--- End diff --
No parentheses for getters
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
