[SPARK-11029] [ML] Add computeCost to KMeansModel in spark.ml

jira: https://issues.apache.org/jira/browse/SPARK-11029

We should add a method analogous to spark.mllib.clustering.KMeansModel.computeCost to spark.ml.clustering.KMeansModel.
This will be a temp fix until we have proper evaluators defined for clustering.

Author: Yuhao Yang <hhbyyh@gmail.com>
Author: yuhaoyang <yuhao@zhanglipings-iMac.local>

Closes #9073 from hhbyyh/computeCost.

mllib/src/main/scala/org/apache/spark/ml/clustering/KMeans.scala

@@ -117,6 +117,18 @@ class KMeansModel private[ml] (
 
   @Since("1.5.0")
   def clusterCenters: Array[Vector] = parentModel.clusterCenters
+
+  /**
+   * Return the K-means cost (sum of squared distances of points to their nearest center) for this
+   * model on the given data.
+   */
+  // TODO: Replace the temp fix when we have proper evaluators defined for clustering.
+  @Since("1.6.0")
+  def computeCost(dataset: DataFrame): Double = {
+    SchemaUtils.checkColumnType(dataset.schema, $(featuresCol), new VectorUDT)
+    val data = dataset.select(col($(featuresCol))).map { case Row(point: Vector) => point }
+    parentModel.computeCost(data)
+  }
 }
 
 /**

mllib/src/test/scala/org/apache/spark/ml/clustering/KMeansSuite.scala

@@ -104,5 +104,6 @@ class KMeansSuite extends SparkFunSuite with MLlibTestSparkContext {
     val clusters = transformed.select(predictionColName).map(_.getInt(0)).distinct().collect().toSet
     assert(clusters.size === k)
     assert(clusters === Set(0, 1, 2, 3, 4))
+    assert(model.computeCost(dataset) < 0.1)
   }
 }