[SPARK-10240] [SPARK-10242] [MLLIB] update since versions in mlilb.random and mllib.stat

The same as #8241 but for `mllib.stat` and `mllib.random`.

cc feynmanliang

Author: Xiangrui Meng <meng@databricks.com>

Closes #8439 from mengxr/SPARK-10242.

(cherry picked from commit c3a54843c0c8a14059da4e6716c1ad45c69bbe6c)
Signed-off-by: Xiangrui Meng <meng@databricks.com>

4 files changed, 117 insertions, 25 deletions

diff --git a/mllib/src/main/scala/org/apache/spark/mllib/random/RandomDataGenerator.scala b/mllib/src/main/scala/org/apache/spark/mllib/random/RandomDataGenerator.scala
index 9349ecaa13..a2d85a68cd 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/random/RandomDataGenerator.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/random/RandomDataGenerator.scala
@@ -20,7 +20,7 @@ package org.apache.spark.mllib.random
 import org.apache.commons.math3.distribution.{ExponentialDistribution,
   GammaDistribution, LogNormalDistribution, PoissonDistribution}
 
-import org.apache.spark.annotation.DeveloperApi
+import org.apache.spark.annotation.{Since, DeveloperApi}
 import org.apache.spark.util.random.{XORShiftRandom, Pseudorandom}
 
 /**
@@ -28,17 +28,20 @@ import org.apache.spark.util.random.{XORShiftRandom, Pseudorandom}
  * Trait for random data generators that generate i.i.d. data.
  */
 @DeveloperApi
+@Since("1.1.0")
 trait RandomDataGenerator[T] extends Pseudorandom with Serializable {
 
   /**
    * Returns an i.i.d. sample as a generic type from an underlying distribution.
    */
+  @Since("1.1.0")
   def nextValue(): T
 
   /**
    * Returns a copy of the RandomDataGenerator with a new instance of the rng object used in the
    * class when applicable for non-locking concurrent usage.
    */
+  @Since("1.1.0")
   def copy(): RandomDataGenerator[T]
 }
 
@@ -47,17 +50,21 @@ trait RandomDataGenerator[T] extends Pseudorandom with Serializable {
  * Generates i.i.d. samples from U[0.0, 1.0]
  */
 @DeveloperApi
+@Since("1.1.0")
 class UniformGenerator extends RandomDataGenerator[Double] {
 
   // XORShiftRandom for better performance. Thread safety isn't necessary here.
   private val random = new XORShiftRandom()
 
+  @Since("1.1.0")
   override def nextValue(): Double = {
     random.nextDouble()
   }
 
+  @Since("1.1.0")
   override def setSeed(seed: Long): Unit = random.setSeed(seed)
 
+  @Since("1.1.0")
   override def copy(): UniformGenerator = new UniformGenerator()
 }
 
@@ -66,17 +73,21 @@ class UniformGenerator extends RandomDataGenerator[Double] {
  * Generates i.i.d. samples from the standard normal distribution.
  */
 @DeveloperApi
+@Since("1.1.0")
 class StandardNormalGenerator extends RandomDataGenerator[Double] {
 
   // XORShiftRandom for better performance. Thread safety isn't necessary here.
   private val random = new XORShiftRandom()
 
+  @Since("1.1.0")
   override def nextValue(): Double = {
       random.nextGaussian()
   }
 
+  @Since("1.1.0")
   override def setSeed(seed: Long): Unit = random.setSeed(seed)
 
+  @Since("1.1.0")
   override def copy(): StandardNormalGenerator = new StandardNormalGenerator()
 }
 
@@ -87,16 +98,21 @@ class StandardNormalGenerator extends RandomDataGenerator[Double] {
  * @param mean mean for the Poisson distribution.
  */
 @DeveloperApi
-class PoissonGenerator(val mean: Double) extends RandomDataGenerator[Double] {
+@Since("1.1.0")
+class PoissonGenerator @Since("1.1.0") (
+    @Since("1.1.0") val mean: Double) extends RandomDataGenerator[Double] {
 
   private val rng = new PoissonDistribution(mean)
 
+  @Since("1.1.0")
   override def nextValue(): Double = rng.sample()
 
+  @Since("1.1.0")
   override def setSeed(seed: Long) {
     rng.reseedRandomGenerator(seed)
   }
 
+  @Since("1.1.0")
   override def copy(): PoissonGenerator = new PoissonGenerator(mean)
 }
 
@@ -107,16 +123,21 @@ class PoissonGenerator(val mean: Double) extends RandomDataGenerator[Double] {
  * @param mean mean for the exponential distribution.
  */
 @DeveloperApi
-class ExponentialGenerator(val mean: Double) extends RandomDataGenerator[Double] {
+@Since("1.3.0")
+class ExponentialGenerator @Since("1.3.0") (
+    @Since("1.3.0") val mean: Double) extends RandomDataGenerator[Double] {
 
   private val rng = new ExponentialDistribution(mean)
 
+  @Since("1.3.0")
   override def nextValue(): Double = rng.sample()
 
+  @Since("1.3.0")
   override def setSeed(seed: Long) {
     rng.reseedRandomGenerator(seed)
   }
 
+  @Since("1.3.0")
   override def copy(): ExponentialGenerator = new ExponentialGenerator(mean)
 }
 
@@ -128,16 +149,22 @@ class ExponentialGenerator(val mean: Double) extends RandomDataGenerator[Double]
  * @param scale scale for the gamma distribution
  */
 @DeveloperApi
-class GammaGenerator(val shape: Double, val scale: Double) extends RandomDataGenerator[Double] {
+@Since("1.3.0")
+class GammaGenerator @Since("1.3.0") (
+    @Since("1.3.0") val shape: Double,
+    @Since("1.3.0") val scale: Double) extends RandomDataGenerator[Double] {
 
   private val rng = new GammaDistribution(shape, scale)
 
+  @Since("1.3.0")
   override def nextValue(): Double = rng.sample()
 
+  @Since("1.3.0")
   override def setSeed(seed: Long) {
     rng.reseedRandomGenerator(seed)
   }
 
+  @Since("1.3.0")
   override def copy(): GammaGenerator = new GammaGenerator(shape, scale)
 }
 
@@ -150,15 +177,21 @@ class GammaGenerator(val shape: Double, val scale: Double) extends RandomDataGen
  * @param std standard deviation for the log normal distribution
  */
 @DeveloperApi
-class LogNormalGenerator(val mean: Double, val std: Double) extends RandomDataGenerator[Double] {
+@Since("1.3.0")
+class LogNormalGenerator @Since("1.3.0") (
+    @Since("1.3.0") val mean: Double,
+    @Since("1.3.0") val std: Double) extends RandomDataGenerator[Double] {
 
   private val rng = new LogNormalDistribution(mean, std)
 
+  @Since("1.3.0")
   override def nextValue(): Double = rng.sample()
 
+  @Since("1.3.0")
   override def setSeed(seed: Long) {
     rng.reseedRandomGenerator(seed)
   }
 
+  @Since("1.3.0")
   override def copy(): LogNormalGenerator = new LogNormalGenerator(mean, std)
 }
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/random/RandomRDDs.scala b/mllib/src/main/scala/org/apache/spark/mllib/random/RandomRDDs.scala
index 174d5e0f6c..4dd5ea214d 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/random/RandomRDDs.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/random/RandomRDDs.scala
@@ -20,7 +20,7 @@ package org.apache.spark.mllib.random
 import scala.reflect.ClassTag
 
 import org.apache.spark.SparkContext
-import org.apache.spark.annotation.{DeveloperApi, Experimental}
+import org.apache.spark.annotation.{DeveloperApi, Experimental, Since}
 import org.apache.spark.api.java.{JavaDoubleRDD, JavaRDD, JavaSparkContext}
 import org.apache.spark.mllib.linalg.Vector
 import org.apache.spark.mllib.rdd.{RandomRDD, RandomVectorRDD}
@@ -32,6 +32,7 @@ import org.apache.spark.util.Utils
  * Generator methods for creating RDDs comprised of `i.i.d.` samples from some distribution.
  */
 @Experimental
+@Since("1.1.0")
 object RandomRDDs {
 
   /**
@@ -46,6 +47,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Double] comprised of `i.i.d.` samples ~ `U(0.0, 1.0)`.
    */
+  @Since("1.1.0")
   def uniformRDD(
       sc: SparkContext,
       size: Long,
@@ -58,6 +60,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#uniformRDD]].
    */
+  @Since("1.1.0")
   def uniformJavaRDD(
       jsc: JavaSparkContext,
       size: Long,
@@ -69,6 +72,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#uniformJavaRDD]] with the default seed.
    */
+  @Since("1.1.0")
   def uniformJavaRDD(jsc: JavaSparkContext, size: Long, numPartitions: Int): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(uniformRDD(jsc.sc, size, numPartitions))
   }
@@ -76,6 +80,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#uniformJavaRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.1.0")
   def uniformJavaRDD(jsc: JavaSparkContext, size: Long): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(uniformRDD(jsc.sc, size))
   }
@@ -92,6 +97,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Double] comprised of `i.i.d.` samples ~ N(0.0, 1.0).
    */
+  @Since("1.1.0")
   def normalRDD(
       sc: SparkContext,
       size: Long,
@@ -104,6 +110,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#normalRDD]].
    */
+  @Since("1.1.0")
   def normalJavaRDD(
       jsc: JavaSparkContext,
       size: Long,
@@ -115,6 +122,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#normalJavaRDD]] with the default seed.
    */
+  @Since("1.1.0")
   def normalJavaRDD(jsc: JavaSparkContext, size: Long, numPartitions: Int): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(normalRDD(jsc.sc, size, numPartitions))
   }
@@ -122,6 +130,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#normalJavaRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.1.0")
   def normalJavaRDD(jsc: JavaSparkContext, size: Long): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(normalRDD(jsc.sc, size))
   }
@@ -137,6 +146,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Double] comprised of `i.i.d.` samples ~ Pois(mean).
    */
+  @Since("1.1.0")
   def poissonRDD(
       sc: SparkContext,
       mean: Double,
@@ -150,6 +160,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#poissonRDD]].
    */
+  @Since("1.1.0")
   def poissonJavaRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -162,6 +173,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#poissonJavaRDD]] with the default seed.
    */
+  @Since("1.1.0")
   def poissonJavaRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -173,6 +185,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#poissonJavaRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.1.0")
   def poissonJavaRDD(jsc: JavaSparkContext, mean: Double, size: Long): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(poissonRDD(jsc.sc, mean, size))
   }
@@ -188,6 +201,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Double] comprised of `i.i.d.` samples ~ Pois(mean).
    */
+  @Since("1.3.0")
   def exponentialRDD(
       sc: SparkContext,
       mean: Double,
@@ -201,6 +215,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#exponentialRDD]].
    */
+  @Since("1.3.0")
   def exponentialJavaRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -213,6 +228,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#exponentialJavaRDD]] with the default seed.
    */
+  @Since("1.3.0")
   def exponentialJavaRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -224,6 +240,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#exponentialJavaRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.3.0")
   def exponentialJavaRDD(jsc: JavaSparkContext, mean: Double, size: Long): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(exponentialRDD(jsc.sc, mean, size))
   }
@@ -240,6 +257,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Double] comprised of `i.i.d.` samples ~ Pois(mean).
    */
+  @Since("1.3.0")
   def gammaRDD(
       sc: SparkContext,
       shape: Double,
@@ -254,6 +272,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#gammaRDD]].
    */
+  @Since("1.3.0")
   def gammaJavaRDD(
       jsc: JavaSparkContext,
       shape: Double,
@@ -267,6 +286,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#gammaJavaRDD]] with the default seed.
    */
+  @Since("1.3.0")
   def gammaJavaRDD(
       jsc: JavaSparkContext,
       shape: Double,
@@ -279,11 +299,12 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#gammaJavaRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.3.0")
   def gammaJavaRDD(
-    jsc: JavaSparkContext,
-    shape: Double,
-    scale: Double,
-    size: Long): JavaDoubleRDD = {
+      jsc: JavaSparkContext,
+      shape: Double,
+      scale: Double,
+      size: Long): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(gammaRDD(jsc.sc, shape, scale, size))
   }
 
@@ -299,6 +320,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Double] comprised of `i.i.d.` samples ~ Pois(mean).
    */
+  @Since("1.3.0")
   def logNormalRDD(
       sc: SparkContext,
       mean: Double,
@@ -313,6 +335,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#logNormalRDD]].
    */
+  @Since("1.3.0")
   def logNormalJavaRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -326,6 +349,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#logNormalJavaRDD]] with the default seed.
    */
+  @Since("1.3.0")
   def logNormalJavaRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -338,11 +362,12 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#logNormalJavaRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.3.0")
   def logNormalJavaRDD(
-    jsc: JavaSparkContext,
-    mean: Double,
-    std: Double,
-    size: Long): JavaDoubleRDD = {
+      jsc: JavaSparkContext,
+      mean: Double,
+      std: Double,
+      size: Long): JavaDoubleRDD = {
     JavaDoubleRDD.fromRDD(logNormalRDD(jsc.sc, mean, std, size))
   }
 
@@ -359,6 +384,7 @@ object RandomRDDs {
    * @return RDD[Double] comprised of `i.i.d.` samples produced by generator.
    */
   @DeveloperApi
+  @Since("1.1.0")
   def randomRDD[T: ClassTag](
       sc: SparkContext,
       generator: RandomDataGenerator[T],
@@ -381,6 +407,7 @@ object RandomRDDs {
    * @param seed Seed for the RNG that generates the seed for the generator in each partition.
    * @return RDD[Vector] with vectors containing i.i.d samples ~ `U(0.0, 1.0)`.
    */
+  @Since("1.1.0")
   def uniformVectorRDD(
       sc: SparkContext,
       numRows: Long,
@@ -394,6 +421,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#uniformVectorRDD]].
    */
+  @Since("1.1.0")
   def uniformJavaVectorRDD(
       jsc: JavaSparkContext,
       numRows: Long,
@@ -406,6 +434,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#uniformJavaVectorRDD]] with the default seed.
    */
+  @Since("1.1.0")
   def uniformJavaVectorRDD(
       jsc: JavaSparkContext,
       numRows: Long,
@@ -417,6 +446,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#uniformJavaVectorRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.1.0")
   def uniformJavaVectorRDD(
       jsc: JavaSparkContext,
       numRows: Long,
@@ -435,6 +465,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Vector] with vectors containing `i.i.d.` samples ~ `N(0.0, 1.0)`.
    */
+  @Since("1.1.0")
   def normalVectorRDD(
       sc: SparkContext,
       numRows: Long,
@@ -448,6 +479,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#normalVectorRDD]].
    */
+  @Since("1.1.0")
   def normalJavaVectorRDD(
       jsc: JavaSparkContext,
       numRows: Long,
@@ -460,6 +492,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#normalJavaVectorRDD]] with the default seed.
    */
+  @Since("1.1.0")
   def normalJavaVectorRDD(
       jsc: JavaSparkContext,
       numRows: Long,
@@ -471,6 +504,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#normalJavaVectorRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.1.0")
   def normalJavaVectorRDD(
       jsc: JavaSparkContext,
       numRows: Long,
@@ -491,6 +525,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Vector] with vectors containing `i.i.d.` samples.
    */
+  @Since("1.3.0")
   def logNormalVectorRDD(
       sc: SparkContext,
       mean: Double,
@@ -507,6 +542,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#logNormalVectorRDD]].
    */
+  @Since("1.3.0")
   def logNormalJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -521,6 +557,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#logNormalJavaVectorRDD]] with the default seed.
    */
+  @Since("1.3.0")
   def logNormalJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -535,6 +572,7 @@ object RandomRDDs {
    * [[RandomRDDs#logNormalJavaVectorRDD]] with the default number of partitions and
    * the default seed.
    */
+  @Since("1.3.0")
   def logNormalJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -556,6 +594,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Vector] with vectors containing `i.i.d.` samples ~ Pois(mean).
    */
+  @Since("1.1.0")
   def poissonVectorRDD(
       sc: SparkContext,
       mean: Double,
@@ -570,6 +609,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#poissonVectorRDD]].
    */
+  @Since("1.1.0")
   def poissonJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -583,6 +623,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#poissonJavaVectorRDD]] with the default seed.
    */
+  @Since("1.1.0")
   def poissonJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -595,6 +636,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#poissonJavaVectorRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.1.0")
   def poissonJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -615,6 +657,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Vector] with vectors containing `i.i.d.` samples ~ Exp(mean).
    */
+  @Since("1.3.0")
   def exponentialVectorRDD(
       sc: SparkContext,
       mean: Double,
@@ -630,6 +673,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#exponentialVectorRDD]].
    */
+  @Since("1.3.0")
   def exponentialJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -643,6 +687,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#exponentialJavaVectorRDD]] with the default seed.
    */
+  @Since("1.3.0")
   def exponentialJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -656,6 +701,7 @@ object RandomRDDs {
    * [[RandomRDDs#exponentialJavaVectorRDD]] with the default number of partitions
    * and the default seed.
    */
+  @Since("1.3.0")
   def exponentialJavaVectorRDD(
       jsc: JavaSparkContext,
       mean: Double,
@@ -678,6 +724,7 @@ object RandomRDDs {
    * @param seed Random seed (default: a random long integer).
    * @return RDD[Vector] with vectors containing `i.i.d.` samples ~ Exp(mean).
    */
+  @Since("1.3.0")
   def gammaVectorRDD(
       sc: SparkContext,
       shape: Double,
@@ -693,6 +740,7 @@ object RandomRDDs {
   /**
    * Java-friendly version of [[RandomRDDs#gammaVectorRDD]].
    */
+  @Since("1.3.0")
   def gammaJavaVectorRDD(
       jsc: JavaSparkContext,
       shape: Double,
@@ -707,6 +755,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#gammaJavaVectorRDD]] with the default seed.
    */
+  @Since("1.3.0")
   def gammaJavaVectorRDD(
       jsc: JavaSparkContext,
       shape: Double,
@@ -720,6 +769,7 @@ object RandomRDDs {
   /**
    * [[RandomRDDs#gammaJavaVectorRDD]] with the default number of partitions and the default seed.
    */
+  @Since("1.3.0")
   def gammaJavaVectorRDD(
       jsc: JavaSparkContext,
       shape: Double,
@@ -744,6 +794,7 @@ object RandomRDDs {
    * @return RDD[Vector] with vectors containing `i.i.d.` samples produced by generator.
    */
   @DeveloperApi
+  @Since("1.1.0")
   def randomVectorRDD(sc: SparkContext,
       generator: RandomDataGenerator[Double],
       numRows: Long,
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/stat/distribution/MultivariateGaussian.scala b/mllib/src/main/scala/org/apache/spark/mllib/stat/distribution/MultivariateGaussian.scala
index bd4d81390b..92a5af708d 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/stat/distribution/MultivariateGaussian.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/stat/distribution/MultivariateGaussian.scala
@@ -35,9 +35,9 @@ import org.apache.spark.mllib.util.MLUtils
  */
 @Since("1.3.0")
 @DeveloperApi
-class MultivariateGaussian (
-    val mu: Vector,
-    val sigma: Matrix) extends Serializable {
+class MultivariateGaussian @Since("1.3.0") (
+    @Since("1.3.0") val mu: Vector,
+    @Since("1.3.0") val sigma: Matrix) extends Serializable {
 
   require(sigma.numCols == sigma.numRows, "Covariance matrix must be square")
   require(mu.size == sigma.numCols, "Mean vector length must match covariance matrix size")
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/stat/test/TestResult.scala b/mllib/src/main/scala/org/apache/spark/mllib/stat/test/TestResult.scala
index f44be13706..d01b3707be 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/stat/test/TestResult.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/stat/test/TestResult.scala
@@ -17,7 +17,7 @@
 
 package org.apache.spark.mllib.stat.test
 
-import org.apache.spark.annotation.Experimental
+import org.apache.spark.annotation.{Experimental, Since}
 
 /**
  * :: Experimental ::
@@ -25,28 +25,33 @@ import org.apache.spark.annotation.Experimental
  * @tparam DF Return type of `degreesOfFreedom`.
  */
 @Experimental
+@Since("1.1.0")
 trait TestResult[DF] {
 
   /**
    * The probability of obtaining a test statistic result at least as extreme as the one that was
    * actually observed, assuming that the null hypothesis is true.
    */
+  @Since("1.1.0")
   def pValue: Double
 
   /**
    * Returns the degree(s) of freedom of the hypothesis test.
    * Return type should be Number(e.g. Int, Double) or tuples of Numbers for toString compatibility.
    */
+  @Since("1.1.0")
   def degreesOfFreedom: DF
 
   /**
    * Test statistic.
    */
+  @Since("1.1.0")
   def statistic: Double
 
   /**
    * Null hypothesis of the test.
    */
+  @Since("1.1.0")
   def nullHypothesis: String
 
   /**
@@ -78,11 +83,12 @@ trait TestResult[DF] {
  * Object containing the test results for the chi-squared hypothesis test.
  */
 @Experimental
+@Since("1.1.0")
 class ChiSqTestResult private[stat] (override val pValue: Double,
-    override val degreesOfFreedom: Int,
-    override val statistic: Double,
-    val method: String,
-    override val nullHypothesis: String) extends TestResult[Int] {
+    @Since("1.1.0") override val degreesOfFreedom: Int,
+    @Since("1.1.0") override val statistic: Double,
+    @Since("1.1.0") val method: String,
+    @Since("1.1.0") override val nullHypothesis: String) extends TestResult[Int] {
 
   override def toString: String = {
     "Chi squared test summary:\n" +
@@ -96,11 +102,13 @@ class ChiSqTestResult private[stat] (override val pValue: Double,
  * Object containing the test results for the Kolmogorov-Smirnov test.
  */
 @Experimental
+@Since("1.5.0")
 class KolmogorovSmirnovTestResult private[stat] (
-    override val pValue: Double,
-    override val statistic: Double,
-    override val nullHypothesis: String) extends TestResult[Int] {
+    @Since("1.5.0") override val pValue: Double,
+    @Since("1.5.0") override val statistic: Double,
+    @Since("1.5.0") override val nullHypothesis: String) extends TestResult[Int] {
 
+  @Since("1.5.0")
   override val degreesOfFreedom = 0
 
   override def toString: String = {