diff --git a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/BinaryClassificationMetrics.scala b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/BinaryClassificationMetrics.scala
index 76ae847921f44..508fe532b1306 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/BinaryClassificationMetrics.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/BinaryClassificationMetrics.scala
@@ -42,11 +42,11 @@ import org.apache.spark.sql.DataFrame
  *                be smaller as a result, meaning there may be an extra sample at
  *                partition boundaries.
  */
-@Since("1.3.0")
+@Since("1.0.0")
 @Experimental
-class BinaryClassificationMetrics(
-    val scoreAndLabels: RDD[(Double, Double)],
-    val numBins: Int) extends Logging {
+class BinaryClassificationMetrics @Since("1.3.0") (
+    @Since("1.3.0") val scoreAndLabels: RDD[(Double, Double)],
+    @Since("1.3.0") val numBins: Int) extends Logging {
 
   require(numBins >= 0, "numBins must be nonnegative")
 
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MulticlassMetrics.scala b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MulticlassMetrics.scala
index 02e89d921033c..00e837661dfc2 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MulticlassMetrics.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MulticlassMetrics.scala
@@ -33,7 +33,7 @@ import org.apache.spark.sql.DataFrame
  */
 @Since("1.1.0")
 @Experimental
-class MulticlassMetrics(predictionAndLabels: RDD[(Double, Double)]) {
+class MulticlassMetrics @Since("1.1.0") (predictionAndLabels: RDD[(Double, Double)]) {
 
   /**
    * An auxiliary constructor taking a DataFrame.
@@ -140,6 +140,7 @@ class MulticlassMetrics(predictionAndLabels: RDD[(Double, Double)]) {
   /**
    * Returns precision
    */
+  @Since("1.1.0")
   lazy val precision: Double = tpByClass.values.sum.toDouble / labelCount
 
   /**
@@ -148,23 +149,27 @@ class MulticlassMetrics(predictionAndLabels: RDD[(Double, Double)]) {
    * because sum of all false positives is equal to sum
    * of all false negatives)
    */
+  @Since("1.1.0")
   lazy val recall: Double = precision
 
   /**
    * Returns f-measure
    * (equals to precision and recall because precision equals recall)
    */
+  @Since("1.1.0")
   lazy val fMeasure: Double = precision
 
   /**
    * Returns weighted true positive rate
    * (equals to precision, recall and f-measure)
    */
+  @Since("1.1.0")
   lazy val weightedTruePositiveRate: Double = weightedRecall
 
   /**
    * Returns weighted false positive rate
    */
+  @Since("1.1.0")
   lazy val weightedFalsePositiveRate: Double = labelCountByClass.map { case (category, count) =>
     falsePositiveRate(category) * count.toDouble / labelCount
   }.sum
@@ -173,6 +178,7 @@ class MulticlassMetrics(predictionAndLabels: RDD[(Double, Double)]) {
    * Returns weighted averaged recall
    * (equals to precision, recall and f-measure)
    */
+  @Since("1.1.0")
   lazy val weightedRecall: Double = labelCountByClass.map { case (category, count) =>
     recall(category) * count.toDouble / labelCount
   }.sum
@@ -180,6 +186,7 @@ class MulticlassMetrics(predictionAndLabels: RDD[(Double, Double)]) {
   /**
    * Returns weighted averaged precision
    */
+  @Since("1.1.0")
   lazy val weightedPrecision: Double = labelCountByClass.map { case (category, count) =>
     precision(category) * count.toDouble / labelCount
   }.sum
@@ -196,6 +203,7 @@ class MulticlassMetrics(predictionAndLabels: RDD[(Double, Double)]) {
   /**
    * Returns weighted averaged f1-measure
    */
+  @Since("1.1.0")
   lazy val weightedFMeasure: Double = labelCountByClass.map { case (category, count) =>
     fMeasure(category, 1.0) * count.toDouble / labelCount
   }.sum
@@ -203,5 +211,6 @@ class MulticlassMetrics(predictionAndLabels: RDD[(Double, Double)]) {
   /**
    * Returns the sequence of labels in ascending order
    */
+  @Since("1.1.0")
   lazy val labels: Array[Double] = tpByClass.keys.toArray.sorted
 }
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MultilabelMetrics.scala b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MultilabelMetrics.scala
index a0a8d9c56847b..c100b3c9ec14a 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MultilabelMetrics.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/MultilabelMetrics.scala
@@ -28,7 +28,7 @@ import org.apache.spark.sql.DataFrame
  * both are non-null Arrays, each with unique elements.
  */
 @Since("1.2.0")
-class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]) {
+class MultilabelMetrics @Since("1.2.0") (predictionAndLabels: RDD[(Array[Double], Array[Double])]) {
 
   /**
    * An auxiliary constructor taking a DataFrame.
@@ -46,6 +46,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
    * Returns subset accuracy
    * (for equal sets of labels)
    */
+  @Since("1.2.0")
   lazy val subsetAccuracy: Double = predictionAndLabels.filter { case (predictions, labels) =>
     predictions.deep == labels.deep
   }.count().toDouble / numDocs
@@ -53,6 +54,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
   /**
    * Returns accuracy
    */
+  @Since("1.2.0")
   lazy val accuracy: Double = predictionAndLabels.map { case (predictions, labels) =>
     labels.intersect(predictions).size.toDouble /
       (labels.size + predictions.size - labels.intersect(predictions).size)}.sum / numDocs
@@ -61,6 +63,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
   /**
    * Returns Hamming-loss
    */
+  @Since("1.2.0")
   lazy val hammingLoss: Double = predictionAndLabels.map { case (predictions, labels) =>
     labels.size + predictions.size - 2 * labels.intersect(predictions).size
   }.sum / (numDocs * numLabels)
@@ -68,6 +71,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
   /**
    * Returns document-based precision averaged by the number of documents
    */
+  @Since("1.2.0")
   lazy val precision: Double = predictionAndLabels.map { case (predictions, labels) =>
     if (predictions.size > 0) {
       predictions.intersect(labels).size.toDouble / predictions.size
@@ -79,6 +83,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
   /**
    * Returns document-based recall averaged by the number of documents
    */
+  @Since("1.2.0")
   lazy val recall: Double = predictionAndLabels.map { case (predictions, labels) =>
     labels.intersect(predictions).size.toDouble / labels.size
   }.sum / numDocs
@@ -86,6 +91,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
   /**
    * Returns document-based f1-measure averaged by the number of documents
    */
+  @Since("1.2.0")
   lazy val f1Measure: Double = predictionAndLabels.map { case (predictions, labels) =>
     2.0 * predictions.intersect(labels).size / (predictions.size + labels.size)
   }.sum / numDocs
@@ -143,6 +149,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
    * Returns micro-averaged label-based precision
    * (equals to micro-averaged document-based precision)
    */
+  @Since("1.2.0")
   lazy val microPrecision: Double = {
     val sumFp = fpPerClass.foldLeft(0L){ case(cum, (_, fp)) => cum + fp}
     sumTp.toDouble / (sumTp + sumFp)
@@ -152,6 +159,7 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
    * Returns micro-averaged label-based recall
    * (equals to micro-averaged document-based recall)
    */
+  @Since("1.2.0")
   lazy val microRecall: Double = {
     val sumFn = fnPerClass.foldLeft(0.0){ case(cum, (_, fn)) => cum + fn}
     sumTp.toDouble / (sumTp + sumFn)
@@ -161,10 +169,12 @@ class MultilabelMetrics(predictionAndLabels: RDD[(Array[Double], Array[Double])]
    * Returns micro-averaged label-based f1-measure
    * (equals to micro-averaged document-based f1-measure)
    */
+  @Since("1.2.0")
   lazy val microF1Measure: Double = 2.0 * sumTp / (2 * sumTp + sumFnClass + sumFpClass)
 
   /**
    * Returns the sequence of labels in ascending order
    */
+  @Since("1.2.0")
   lazy val labels: Array[Double] = tpPerClass.keys.toArray.sorted
 }
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/RegressionMetrics.scala b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/RegressionMetrics.scala
index 36a6c357c3897..799ebb980ef01 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/evaluation/RegressionMetrics.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/evaluation/RegressionMetrics.scala
@@ -32,7 +32,8 @@ import org.apache.spark.sql.DataFrame
  */
 @Since("1.2.0")
 @Experimental
-class RegressionMetrics(predictionAndObservations: RDD[(Double, Double)]) extends Logging {
+class RegressionMetrics @Since("1.2.0") (
+    predictionAndObservations: RDD[(Double, Double)]) extends Logging {
 
   /**
    * An auxiliary constructor taking a DataFrame.