Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/3022#discussion_r21860754
 Diff: mllib/src/main/scala/org/apache/spark/mllib/clustering/GaussianMixtureModelEM.scala

@@ 0,0 +1,234 @@
+/*
+ * 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/LICENSE2.0
+ *
+ * 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 => BreezeVector, DenseMatrix => BreezeMatrix}
+import breeze.linalg.Transpose
+
+import org.apache.spark.rdd.RDD
+import org.apache.spark.mllib.linalg.{Matrices, Vector, Vectors}
+import org.apache.spark.mllib.stat.impl.MultivariateGaussian
+import org.apache.spark.{Accumulator, AccumulatorParam, SparkContext}
+import org.apache.spark.SparkContext.DoubleAccumulatorParam
+
+/**
+ * This class performs expectation maximization for multivariate Gaussian
+ * Mixture Models (GMMs). A GMM represents a composite distribution of
+ * independent Gaussian distributions with associated "mixing" weights
+ * specifying each's contribution to the composite.
+ *
+ * Given a set of sample points, this class will maximize the loglikelihood
+ * for a mixture of k Gaussians, iterating until the loglikelihood changes by
+ * less than convergenceTol, or until it has reached the max number of iterations.
+ * While this process is generally guaranteed to converge, it is not guaranteed
+ * to find a global optimum.
+ *
+ * @param k The number of independent Gaussians in the mixture model
+ * @param convergenceTol The maximum change in loglikelihood at which convergence
+ * is considered to have occurred.
+ * @param maxIterations The maximum number of iterations to perform
+ */
+class GaussianMixtureModelEM private (
+ private var k: Int,
+ private var convergenceTol: Double,
+ private var maxIterations: Int) extends Serializable {
+
+ // Type aliases for convenience
+ private type DenseDoubleVector = BreezeVector[Double]
+ private type DenseDoubleMatrix = BreezeMatrix[Double]
+
+ // number of samples per cluster to use when initializing Gaussians
+ private val nSamples = 5
+
+ /** A default instance, 2 Gaussians, 100 iterations, 0.01 loglikelihood threshold
*/
+ def this() = this(2, 0.01, 100)
+
+ /** Set the number of Gaussians in the mixture model. Default: 2 */
+ def setK(k: Int): this.type = {
+ this.k = k
+ this
+ }
+
+ /** Return the number of Gaussians in the mixture model */
+ def getK: Int = k
+
+ /** Set the maximum number of iterations to run. Default: 100 */
+ def setMaxIterations(maxIterations: Int): this.type = {
+ this.maxIterations = maxIterations
+ this
+ }
+
+ /** Return the maximum number of iterations to run */
+ def getMaxIterations: Int = maxIterations
+
+ /**
+ * Set the largest change in loglikelihood at which convergence is
+ * considered to have occurred.
+ */
+ def setConvergenceTol(convergenceTol: Double): this.type = {
+ this.convergenceTol = convergenceTol
+ this
+ }
+
+ /** Return the largest change in loglikelihood at which convergence is
+ * considered to have occurred.
+ */
+ def getConvergenceTol: Double = convergenceTol
+
+ /** Machine precision value used to ensure matrix conditioning */
+ private val eps = math.pow(2.0, 52)
+
+ /** Perform expectation maximization */
+ def run(data: RDD[Vector]): GaussianMixtureModel = {
+ val ctx = data.sparkContext
+
+ // we will operate on the data as breeze data
+ val breezeData = data.map( u => u.toBreeze.toDenseVector ).cache()
+
+ // Get length of the input vectors
+ val d = breezeData.first.length
+
+ // For each Gaussian, we will initialize the mean as the average
+ // of some random samples from the data
+ val samples = breezeData.takeSample(true, k * nSamples, scala.util.Random.nextInt)
+
+ // gaussians will be array of (weight, mean, covariance) tuples
+ // we start with uniform weights, a random mean from the data, and
+ // diagonal covariance matrices using component variances
+ // derived from the samples
+ var gaussians = (0 until k).map{ i => (1.0 / k,
 End diff 
I would format as:
```
var gaussians = (0 until k).map{ i =>
(1.0 / k,
vectorMean(samples.slice(i * nSamples, (i + 1) * nSamples)),
initCovariance(samples.slice(i * nSamples, (i + 1) * nSamples)))
}.toArray
```
(indentation + ending the first line with "=>")

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 infrastructure@apache.org or file a JIRA ticket
with INFRA.


To unsubscribe, email: reviewsunsubscribe@spark.apache.org
For additional commands, email: reviewshelp@spark.apache.org
