commit dd8514fa2059a695143073f852b1abee50e522bd
author Xusen Yin <yinxusen@gmail.com> Tue Aug 02 07:28:46 2016 -0700
committer Yanbo Liang <ybliang8@gmail.com> Tue Aug 02 07:28:46 2016 -0700
tree e9d1aa02847ead31faa7a3a4a06d6aa08071056b
parent d9e0919d30e9f79a0eb1ceb8d1b5e9fc58cf085e

[SPARK-16558][EXAMPLES][MLLIB] examples/mllib/LDAExample should use MLVector instead of MLlib Vector

## What changes were proposed in this pull request?

mllib.LDAExample uses ML pipeline and MLlib LDA algorithm. The former transforms original data into MLVector format, while the latter uses MLlibVector format.

## How was this patch tested?

Test manually.

Author: Xusen Yin <yinxusen@gmail.com>

Closes #14212 from yinxusen/SPARK-16558.

examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala

diff --git a/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala b/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala
index 7e50b12..b923e62 100644
--- a/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala
@@ -24,8 +24,9 @@
 import org.apache.spark.{SparkConf, SparkContext}
 import org.apache.spark.ml.Pipeline
 import org.apache.spark.ml.feature.{CountVectorizer, CountVectorizerModel, RegexTokenizer, StopWordsRemover}
+import org.apache.spark.ml.linalg.{Vector => MLVector}
 import org.apache.spark.mllib.clustering.{DistributedLDAModel, EMLDAOptimizer, LDA, OnlineLDAOptimizer}
-import org.apache.spark.mllib.linalg.Vector
+import org.apache.spark.mllib.linalg.{Vector, Vectors}
 import org.apache.spark.rdd.RDD
 import org.apache.spark.sql.{Row, SparkSession}
 
@@ -223,7 +224,7 @@
     val documents = model.transform(df)
       .select("features")
       .rdd
-      .map { case Row(features: Vector) => features }
+      .map { case Row(features: MLVector) => Vectors.fromML(features) }
       .zipWithIndex()
       .map(_.swap)