make depth default to 2

Author: mengxr

mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/RowMatrix.scala

@@ -92,9 +92,7 @@ class RowMatrix(
             s"Do not support vector operation from type ${rBrz.getClass.getName}.")
         }
         U
-      },
-      combOp = (U1, U2) => U1 += U2,
-      depth = 2)
+      }, combOp = (U1, U2) => U1 += U2)
   }
 
   /**
@@ -109,7 +107,7 @@ class RowMatrix(
       seqOp = (U, v) => {
         RowMatrix.dspr(1.0, v, U.data)
         U
-      }, combOp = (U1, U2) => U1 += U2, depth = 2)
+      }, combOp = (U1, U2) => U1 += U2)
 
     RowMatrix.triuToFull(n, GU.data)
   }
@@ -292,8 +290,8 @@ class RowMatrix(
     val (m, mean) = rows.treeAggregate[(Long, BDV[Double])]((0L, BDV.zeros[Double](n)))(
       seqOp = (s: (Long, BDV[Double]), v: Vector) => (s._1 + 1L, s._2 += v.toBreeze),
       combOp = (s1: (Long, BDV[Double]), s2: (Long, BDV[Double])) =>
-        (s1._1 + s2._1, s1._2 += s2._2),
-      depth = 2)
+        (s1._1 + s2._1, s1._2 += s2._2)
+    )
 
     updateNumRows(m)
 
@@ -355,8 +353,7 @@ class RowMatrix(
   def computeColumnSummaryStatistics(): MultivariateStatisticalSummary = {
     val summary = rows.treeAggregate(new MultivariateOnlineSummarizer)(
       (aggregator, data) => aggregator.add(data),
-      (aggregator1, aggregator2) => aggregator1.merge(aggregator2),
-      depth = 2)
+      (aggregator1, aggregator2) => aggregator1.merge(aggregator2))
     updateNumRows(summary.count)
     summary
   }

mllib/src/main/scala/org/apache/spark/mllib/optimization/GradientDescent.scala

@@ -185,7 +185,7 @@ object GradientDescent extends Logging {
           },
           combOp = (c1, c2) => (c1, c2) match { case ((grad1, loss1), (grad2, loss2)) =>
             (grad1 += grad2, loss1 + loss2)
-          }, depth = 2)
+          })
 
       /**
        * NOTE(Xinghao): lossSum is computed using the weights from the previous iteration

mllib/src/main/scala/org/apache/spark/mllib/optimization/LBFGS.scala

@@ -208,7 +208,7 @@ object LBFGS extends Logging {
           },
           combOp = (c1, c2) => (c1, c2) match { case ((grad1, loss1), (grad2, loss2)) =>
             (grad1 += grad2, loss1 + loss2)
-          }, depth = 2)
+          })
 
       /**
        * regVal is sum of weight squares if it's L2 updater;

mllib/src/main/scala/org/apache/spark/mllib/rdd/RDDFunctions.scala

@@ -49,11 +49,12 @@ class RDDFunctions[T: ClassTag](self: RDD[T]) {
   }
 
   /**
-   * Reduces the elements of this RDD in a tree pattern.
-   * @param depth suggested depth of the tree
+   * Reduces the elements of this RDD in a multi-level tree pattern.
+   *
+   * @param depth suggested depth of the tree (default: 2)
    * @see [[org.apache.spark.rdd.RDD#reduce]]
    */
-  def treeReduce(f: (T, T) => T, depth: Int): T = {
+  def treeReduce(f: (T, T) => T, depth: Int = 2): T = {
     require(depth >= 1, s"Depth must be greater than or equal to 1 but got $depth.")
     val cleanF = self.context.clean(f)
     val reducePartition: Iterator[T] => Option[T] = iter => {
@@ -80,14 +81,15 @@ class RDDFunctions[T: ClassTag](self: RDD[T]) {
   }
 
   /**
-   * Aggregates the elements of this RDD in a tree pattern.
-   * @param depth suggested depth of the tree
+   * Aggregates the elements of this RDD in a multi-level tree pattern.
+   *
+   * @param depth suggested depth of the tree (default: 2)
    * @see [[org.apache.spark.rdd.RDD#aggregate]]
    */
   def treeAggregate[U: ClassTag](zeroValue: U)(
       seqOp: (U, T) => U,
       combOp: (U, U) => U,
-      depth: Int): U = {
+      depth: Int = 2): U = {
     require(depth >= 1, s"Depth must be greater than or equal to 1 but got $depth.")
     if (self.partitions.size == 0) {
       return Utils.clone(zeroValue, self.context.env.closureSerializer.newInstance())