[SPARK-2514] [mllib] Random RDD generator

Utilities for generating random RDDs.

RandomRDD and RandomVectorRDD are created instead of using `sc.parallelize(range:Range)` because `Range` objects in Scala can only have `size <= Int.MaxValue`.

The object `RandomRDDGenerators` can be transformed into a generator class to reduce the number of auxiliary methods for optional arguments.

Author: Doris Xin <[email protected]>

Closes apache#1520 from dorx/randomRDD and squashes the following commits:

01121ac [Doris Xin] reviewer comments
6bf27d8 [Doris Xin] Merge branch 'master' into randomRDD
a8ea92d [Doris Xin] Reviewer comments
063ea0b [Doris Xin] Merge branch 'master' into randomRDD
aec68eb [Doris Xin] newline
bc90234 [Doris Xin] units passed.
d56cacb [Doris Xin] impl with RandomRDD
92d6f1c [Doris Xin] solution for Cloneable
df5bcff [Doris Xin] Merge branch 'generator' into randomRDD
f46d928 [Doris Xin] WIP
49ed20d [Doris Xin] alternative poisson distribution generator
7cb0e40 [Doris Xin] fix for data inconsistency
8881444 [Doris Xin] RandomRDDGenerator: initial design

mllib/src/main/scala/org/apache/spark/mllib/random/DistributionGenerator.scala

@@ -0,0 +1,101 @@
+/*
+ * 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.random
+
+import cern.jet.random.Poisson
+import cern.jet.random.engine.DRand
+
+import org.apache.spark.annotation.Experimental
+import org.apache.spark.util.random.{XORShiftRandom, Pseudorandom}
+
+/**
+ * :: Experimental ::
+ * Trait for random number generators that generate i.i.d. values from a distribution.
+ */
+@Experimental
+trait DistributionGenerator extends Pseudorandom with Serializable {
+
+  /**
+   * Returns an i.i.d. sample as a Double from an underlying distribution.
+   */
+  def nextValue(): Double
+
+  /**
+   * Returns a copy of the DistributionGenerator with a new instance of the rng object used in the
+   * class when applicable for non-locking concurrent usage.
+   */
+  def copy(): DistributionGenerator
+}
+
+/**
+ * :: Experimental ::
+ * Generates i.i.d. samples from U[0.0, 1.0]
+ */
+@Experimental
+class UniformGenerator extends DistributionGenerator {
+
+  // XORShiftRandom for better performance. Thread safety isn't necessary here.
+  private val random = new XORShiftRandom()
+
+  override def nextValue(): Double = {
+    random.nextDouble()
+  }
+
+  override def setSeed(seed: Long) = random.setSeed(seed)
+
+  override def copy(): UniformGenerator = new UniformGenerator()
+}
+
+/**
+ * :: Experimental ::
+ * Generates i.i.d. samples from the standard normal distribution.
+ */
+@Experimental
+class StandardNormalGenerator extends DistributionGenerator {
+
+  // XORShiftRandom for better performance. Thread safety isn't necessary here.
+  private val random = new XORShiftRandom()
+
+  override def nextValue(): Double = {
+      random.nextGaussian()
+  }
+
+  override def setSeed(seed: Long) = random.setSeed(seed)
+
+  override def copy(): StandardNormalGenerator = new StandardNormalGenerator()
+}
+
+/**
+ * :: Experimental ::
+ * Generates i.i.d. samples from the Poisson distribution with the given mean.
+ *
+ * @param mean mean for the Poisson distribution.
+ */
+@Experimental
+class PoissonGenerator(val mean: Double) extends DistributionGenerator {
+
+  private var rng = new Poisson(mean, new DRand)
+
+  override def nextValue(): Double = rng.nextDouble()
+
+  override def setSeed(seed: Long) {
+    rng = new Poisson(mean, new DRand(seed.toInt))
+  }
+
+  override def copy(): PoissonGenerator = new PoissonGenerator(mean)
+}