use some implicit

Author: DB Tsai

examples/src/main/scala/org/apache/spark/examples/ml/DataFrameExample.scala

@@ -26,7 +26,7 @@ import scopt.OptionParser
 import org.apache.spark.{SparkConf, SparkContext}
 import org.apache.spark.examples.mllib.AbstractParams
 import org.apache.spark.ml.linalg.Vector
-import org.apache.spark.mllib.linalg.{Vectors => OldVectors}
+import org.apache.spark.mllib.linalg.VectorImplicits._
 import org.apache.spark.mllib.stat.MultivariateOnlineSummarizer
 import org.apache.spark.sql.{DataFrame, Row, SQLContext}
 
@@ -82,7 +82,7 @@ object DataFrameExample {
     // Convert features column to an RDD of vectors.
     val features = df.select("features").rdd.map { case Row(v: Vector) => v }
     val featureSummary = features.aggregate(new MultivariateOnlineSummarizer())(
-      (summary, feat) => summary.add(OldVectors.fromML(feat)),
+      (summary, feat) => summary.add(feat),
       (sum1, sum2) => sum1.merge(sum2))
     println(s"Selected features column with average values:\n ${featureSummary.mean.toString}")
 

mllib/src/main/scala/org/apache/spark/ml/ann/Layer.scala

@@ -23,6 +23,7 @@ import breeze.linalg.{*, axpy => Baxpy, DenseMatrix => BDM, DenseVector => BDV,
 
 import org.apache.spark.ml.linalg.{Vector, Vectors}
 import org.apache.spark.mllib.linalg.{Vector => OldVector, Vectors => OldVectors}
+import org.apache.spark.mllib.linalg.VectorImplicits._
 import org.apache.spark.mllib.optimization._
 import org.apache.spark.rdd.RDD
 import org.apache.spark.util.random.XORShiftRandom
@@ -583,9 +584,9 @@ private[ann] class ANNGradient(topology: Topology, dataStacker: DataStacker) ext
     label: Double,
     weights: OldVector,
     cumGradient: OldVector): Double = {
-    val (input, target, realBatchSize) = dataStacker.unstack(data.asML)
-    val model = topology.model(weights.asML)
-    model.computeGradient(input, target, cumGradient.asML, realBatchSize)
+    val (input, target, realBatchSize) = dataStacker.unstack(data)
+    val model = topology.model(weights)
+    model.computeGradient(input, target, cumGradient, realBatchSize)
   }
 }
 
@@ -809,7 +810,7 @@ private[ml] class FeedForwardTrainer(
     // TODO: deprecate standard optimizer because it needs Vector
     val newWeights = optimizer.optimize(dataStacker.stack(data).map { v =>
       (v._1, OldVectors.fromML(v._2))
-    }, OldVectors.fromML(w)).asML
+    }, w)
     topology.model(newWeights)
   }
 

mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala

@@ -33,7 +33,7 @@ import org.apache.spark.ml.param._
 import org.apache.spark.ml.param.shared._
 import org.apache.spark.ml.util._
 import org.apache.spark.mllib.evaluation.BinaryClassificationMetrics
-import org.apache.spark.mllib.linalg.{Vectors => OldVectors}
+import org.apache.spark.mllib.linalg.VectorImplicits._
 import org.apache.spark.mllib.stat.MultivariateOnlineSummarizer
 import org.apache.spark.mllib.util.MLUtils
 import org.apache.spark.rdd.RDD
@@ -280,8 +280,7 @@ class LogisticRegression @Since("1.2.0") (
     val (summarizer, labelSummarizer) = {
       val seqOp = (c: (MultivariateOnlineSummarizer, MultiClassSummarizer),
         instance: Instance) =>
-          (c._1.add(OldVectors.fromML(instance.features), instance.weight),
-           c._2.add(instance.label, instance.weight))
+          (c._1.add(instance.features, instance.weight), c._2.add(instance.label, instance.weight))
 
       val combOp = (c1: (MultivariateOnlineSummarizer, MultiClassSummarizer),
         c2: (MultivariateOnlineSummarizer, MultiClassSummarizer)) =>

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

@@ -33,6 +33,7 @@ import org.apache.spark.mllib.clustering.{DistributedLDAModel => OldDistributedL
 import org.apache.spark.mllib.impl.PeriodicCheckpointer
 import org.apache.spark.mllib.linalg.{Matrices => OldMatrices, Vector => OldVector,
   Vectors => OldVectors}
+import org.apache.spark.mllib.linalg.VectorImplicits._
 import org.apache.spark.rdd.RDD
 import org.apache.spark.sql.{DataFrame, Dataset, Row, SparkSession}
 import org.apache.spark.sql.functions.{col, monotonicallyIncreasingId, udf}
@@ -427,7 +428,7 @@ sealed abstract class LDAModel private[ml] (
    * then this returns the fixed (given) value for the [[docConcentration]] parameter.
    */
   @Since("1.6.0")
-  def estimatedDocConcentration: Vector = getModel.docConcentration.asML
+  def estimatedDocConcentration: Vector = getModel.docConcentration
 
   /**
    * Inferred topics, where each topic is represented by a distribution over terms.
@@ -577,7 +578,7 @@ object LocalLDAModel extends MLReadable[LocalLDAModel] {
       val topicConcentration = data.getAs[Double](3)
       val gammaShape = data.getAs[Double](4)
       val oldModel = new OldLocalLDAModel(OldMatrices.fromML(topicsMatrix),
-        OldVectors.fromML(docConcentration), topicConcentration, gammaShape)
+        docConcentration, topicConcentration, gammaShape)
       val model = new LocalLDAModel(metadata.uid, vocabSize, oldModel, sparkSession)
       DefaultParamsReader.getAndSetParams(model, metadata)
       model
@@ -846,7 +847,7 @@ class LDA @Since("1.6.0") (
     transformSchema(dataset.schema, logging = true)
     val oldLDA = new OldLDA()
       .setK($(k))
-      .setDocConcentration(OldVectors.fromML(getOldDocConcentration))
+      .setDocConcentration(getOldDocConcentration)
       .setTopicConcentration(getOldTopicConcentration)
       .setMaxIterations($(maxIter))
       .setSeed($(seed))

mllib/src/main/scala/org/apache/spark/ml/feature/ElementwiseProduct.scala

@@ -24,6 +24,7 @@ import org.apache.spark.ml.param.Param
 import org.apache.spark.ml.util.{DefaultParamsReadable, DefaultParamsWritable, Identifiable}
 import org.apache.spark.mllib.feature
 import org.apache.spark.mllib.linalg.{Vectors => OldVectors}
+import org.apache.spark.mllib.linalg.VectorImplicits._
 import org.apache.spark.sql.types.DataType
 
 /**
@@ -52,8 +53,8 @@ class ElementwiseProduct(override val uid: String)
 
   override protected def createTransformFunc: Vector => Vector = {
     require(params.contains(scalingVec), s"transformation requires a weight vector")
-    val elemScaler = new feature.ElementwiseProduct(OldVectors.fromML($(scalingVec)))
-    vector => elemScaler.transform(OldVectors.fromML(vector)).asML
+    val elemScaler = new feature.ElementwiseProduct($(scalingVec))
+    v => elemScaler.transform(v)
   }
 
   override protected def outputDataType: DataType = new VectorUDT()