Qiita Teams that are logged in
You are not logged in to any team

Log in to Qiita Team
Community
OrganizationAdvent CalendarQiitadon (β)
Service
Qiita JobsQiita ZineQiita Blog
Help us understand the problem. What is going on with this article?

Spark MLlibで相関係数を算出してみる。

More than 5 years have passed since last update.

環境

  • Mac OS X Version 10.9.5
  • Scala 2.10.4
  • Spark 1.1.0
  • Spark MLlib 1.1.0
  • sbt 0.13.1

準備

/root/to/project/path
   |-- build.sbt
   |-- src
   |    |-- main
   |    |    |-- scala
   |    |    |    |-- Stoppable.scala
   |    |-- test
   |    |    |-- scala
   |    |    |    |-- SparkMLlibCorrelationSpec.scala
build.sbt
name := "Spark MLlib examples"                                                                                                                                                    

version := "0.0.1-SNAPSHOT"

scalaVersion := "2.10.4"

scalacOptions ++= Seq("-Xlint", "-deprecation", "-unchecked", "-feature", "-Xelide-below", "ALL")

libraryDependencies ++= Seq(
  "org.apache.spark" %% "spark-core" % "1.1.0",
  "org.apache.spark" %% "spark-mllib" % "1.1.0",
  "org.specs2" %% "specs2" % "2.4.1"
)

実装

src/test/scala/SparkMLlibCorrelationSpec.scala
import commons._                                                                                                                                                                  

import org.apache.spark._
import org.apache.spark.mllib.stat._
import org.apache.spark.mllib.stat.correlation._
import org.apache.spark.SparkContext._

import org.specs2._

class SparkMLlibCorrelationSpec extends Specification with Stoppable { def is = s2"""

  Spark MLlib

  Correlation
    correlation1                               $correlation1
  """

  // 気温とビールの出荷数
  val xData = Array(12.1, 15.3, 18.6, 21.7, 26.1, 32.1)
  val yData = Array(45.0, 520.0, 2864.0, 6874.0, 25487.0, 102870.0)

  var retCorrelation1: Double = _ 
  using(new SparkContext("local[1]", "SparkMLlibCorrelationSpec", System.getenv("SPARK_HOME"))) { sc =>
    val x = sc.parallelize(xData)
    val y = sc.parallelize(yData)
    retCorrelation1 = Statistics.corr(x, y, "pearson")
  }
  // この相関係数を算出しているページ:http://dekiru.net/article/4576/
  def correlation1 = retCorrelation1 must_== 0.8633475827899126
}
src/main/scala/Stoppable.scala
package commons                                                                                                                                                                   

import scala.language.reflectiveCalls

trait Stoppable {
  type T = { def stop(): Unit }
  def using[A <: T, B](resource: A)(f: A => B) = try {
    f(resource)
  } finally {
    resource.stop()
  }
}

実行

$ sbt '~test-only SparkMLlibCorrelationSpec'

参考

http://spark.apache.org/docs/1.1.0/mllib-statistics.html#correlations

Why not register and get more from Qiita?
  1. We will deliver articles that match you
    By following users and tags, you can catch up information on technical fields that you are interested in as a whole
  2. you can read useful information later efficiently
    By "stocking" the articles you like, you can search right away