diff options
author | Jakub Dubovsky <dubovsky@avast.com> | 2014-12-30 14:19:07 -0800 |
---|---|---|
committer | Xiangrui Meng <meng@databricks.com> | 2014-12-30 14:19:07 -0800 |
commit | 0f31992c61f6662e5347745f6a1ac272a5fd63c9 (patch) | |
tree | ab53848611b13d4e62bd53b2399feb1eca74f0de | |
parent | b239ea1c31aeaa752d5dc8f45423df1f8c0924ca (diff) | |
download | spark-0f31992c61f6662e5347745f6a1ac272a5fd63c9.tar.gz spark-0f31992c61f6662e5347745f6a1ac272a5fd63c9.tar.bz2 spark-0f31992c61f6662e5347745f6a1ac272a5fd63c9.zip |
[Spark-4995] Replace Vector.toBreeze.activeIterator with foreachActive
New foreachActive method of vector was introduced by SPARK-4431 as more efficient alternative to vector.toBreeze.activeIterator. There are some parts of codebase where it was not yet replaced.
dbtsai
Author: Jakub Dubovsky <dubovsky@avast.com>
Closes #3846 from james64/SPARK-4995-foreachActive and squashes the following commits:
3eb7e37 [Jakub Dubovsky] Scalastyle fix
32fe6c6 [Jakub Dubovsky] activeIterator removed - IndexedRowMatrix.toBreeze
47a4777 [Jakub Dubovsky] activeIterator removed in RowMatrix.toBreeze
90a7d98 [Jakub Dubovsky] activeIterator removed in MLUtils.saveAsLibSVMFile
3 files changed, 8 insertions, 6 deletions
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/IndexedRowMatrix.scala b/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/IndexedRowMatrix.scala index 5c1acca0ec..36d8cadd2b 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/IndexedRowMatrix.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/IndexedRowMatrix.scala @@ -142,7 +142,7 @@ class IndexedRowMatrix( val mat = BDM.zeros[Double](m, n) rows.collect().foreach { case IndexedRow(rowIndex, vector) => val i = rowIndex.toInt - vector.toBreeze.activeIterator.foreach { case (j, v) => + vector.foreachActive { case (j, v) => mat(i, j) = v } } diff --git a/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/RowMatrix.scala b/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/RowMatrix.scala index 10a515af88..a3fca53929 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/RowMatrix.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/RowMatrix.scala @@ -588,8 +588,8 @@ class RowMatrix( val n = numCols().toInt val mat = BDM.zeros[Double](m, n) var i = 0 - rows.collect().foreach { v => - v.toBreeze.activeIterator.foreach { case (j, v) => + rows.collect().foreach { vector => + vector.foreachActive { case (j, v) => mat(i, j) = v } i += 1 diff --git a/mllib/src/main/scala/org/apache/spark/mllib/util/MLUtils.scala b/mllib/src/main/scala/org/apache/spark/mllib/util/MLUtils.scala index 1d07b5dab8..da0da0a168 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/util/MLUtils.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/util/MLUtils.scala @@ -154,10 +154,12 @@ object MLUtils { def saveAsLibSVMFile(data: RDD[LabeledPoint], dir: String) { // TODO: allow to specify label precision and feature precision. val dataStr = data.map { case LabeledPoint(label, features) => - val featureStrings = features.toBreeze.activeIterator.map { case (i, v) => - s"${i + 1}:$v" + val sb = new StringBuilder(label.toString) + features.foreachActive { case (i, v) => + sb += ' ' + sb ++= s"${i + 1}:$v" } - (Iterator(label) ++ featureStrings).mkString(" ") + sb.mkString } dataStr.saveAsTextFile(dir) } |