remove computeStat in MLUtils

Author: yinxusen

mllib/src/main/scala/org/apache/spark/mllib/util/MLUtils.scala

@@ -17,14 +17,13 @@
 
 package org.apache.spark.mllib.util
 
-import breeze.linalg.{Vector => BV, DenseVector => BDV, SparseVector => BSV,
-  squaredDistance => breezeSquaredDistance}
+import breeze.linalg.{Vector => BV, SparseVector => BSV, squaredDistance => breezeSquaredDistance}
 
 import org.apache.spark.annotation.Experimental
 import org.apache.spark.SparkContext
 import org.apache.spark.rdd.RDD
 import org.apache.spark.mllib.regression.LabeledPoint
-import org.apache.spark.mllib.linalg.{Vector, Vectors}
+import org.apache.spark.mllib.linalg.Vectors
 
 /**
  * Helper methods to load, save and pre-process data used in ML Lib.
@@ -170,7 +169,7 @@ object MLUtils {
    *     xColMean - Row vector with mean for every column (or feature) of the input data
    *     xColSd - Row vector standard deviation for every column (or feature) of the input data.
    */
-  private[mllib] def computeStats(
+  def computeStats(
       data: RDD[LabeledPoint],
       numFeatures: Int,
       numExamples: Long): (Double, Vector, Vector) = {

mllib/src/test/scala/org/apache/spark/mllib/util/MLUtilsSuite.scala

@@ -27,7 +27,6 @@ import com.google.common.base.Charsets
 import com.google.common.io.Files
 
 import org.apache.spark.mllib.linalg.Vectors
-import org.apache.spark.mllib.regression.LabeledPoint
 import org.apache.spark.mllib.util.MLUtils._
 
 class MLUtilsSuite extends FunSuite with LocalSparkContext {
@@ -56,18 +55,6 @@ class MLUtilsSuite extends FunSuite with LocalSparkContext {
     }
   }
 
-  test("compute stats") {
-    val data = Seq.fill(3)(Seq(
-      LabeledPoint(1.0, Vectors.dense(1.0, 2.0, 3.0)),
-      LabeledPoint(0.0, Vectors.dense(3.0, 4.0, 5.0))
-    )).flatten
-    val rdd = sc.parallelize(data, 2)
-    val (meanLabel, mean, std) = MLUtils.computeStats(rdd, 3, 6)
-    assert(meanLabel === 0.5)
-    assert(mean === Vectors.dense(2.0, 3.0, 4.0))
-    assert(std === Vectors.dense(1.0, 1.0, 1.0))
-  }
-
   test("loadLibSVMData") {
     val lines =
       """