Yu Ishikawa yu-iskw

Source Code by scalatest

class CheckPerformanceSuite extends FunSuite {
  test("check the performance") {
    var start = System.currentTimeMillis()
    val vectors = (1 to 3000000).map { i => Vectors.dense(Math.random(), -1 * Math.random(), Math.random())}
    var end = System.currentTimeMillis()
    println(s"Time for Creating: ${end - start}")

Test Code

import breeze.numerics.abs
import org.apache.spark.mllib.linalg.Vectors
import org.scalatest.FunSuite

class CheckPerformanceSuite extends FunSuite {
  test("check the performance") {
    var start = System.currentTimeMillis()

Question

What is the problem of my SparkR's code?
How do we deal with array:c() and list:list() for an arguments to call Java method?

Diff

https://github.com/yu-iskw/spark/commit/e2852699b5b76c11d70f9c061729231f72f094a9

Error

if (!require(pkg)){ 
        install.packages(pkg) 
}

if ("somepackage" %in% row.names(installed.packages())  == FALSE) 
install.packages("somepackage")

	#!/bin/sh

	# one way (older scala version will be installed)
	# sudo apt-get install scala

	#2nd way
	sudo apt-get remove scala-library scala
	wget www.scala-lang.org/files/archive/scala-2.10.3.deb
	sudo dpkg -i scala-2.10.3.deb
	sudo apt-get update

	import breeze.generic.UFunc
	import breeze.linalg.{sum, DenseVector => DBV, Vector => BV}
	import breeze.macros.expand

	object distance extends UFunc {

	val DEFAULT_METHOD = "euclidean"

	@expand
	@expand.valify

	package breeze.linalg.functions

	import breeze.generic.UFunc
	import breeze.linalg.{SparseVector, DenseVector}
	import breeze.numerics.sqrt

	/**
	* A weighted Euclidean distance function implementation
	*/
	object weightedEuclideanDistance extends UFunc {

	/**
	* bisecting <master> <input> <nNodes> <subIterations>
	*
	* divisive hierarchical clustering using bisecting k-means
	* assumes input is a text file, each row is a data point
	* given as numbers separated by spaces
	*
	*/

	import org.apache.spark.SparkContext

	using Distributions

	export Measure, DiscreteMeasure, Estimator, EstimatorQuality,
	blb, randsubset

	abstract Measure

	type DiscreteMeasure{S<:Number,T<:Number} <: Measure
	points :: Vector{S}
	weights :: Vector{T}

	merge(x=data.frame.x, y=data.frame.y
	, by.x=c("clm1", "clm2")
	, by.y=c("clm1", "clm2")
	)