Machine Learning with Spark - Second Edition

MaxBins

Next we look at how changing the maximum number of bins affects the RMSLE values.

Scala

Let us look at the sample implementation in Scala. We will calculate the RMSLE value for maximum number of bins: 10, 16, 32, and 64.

object GradientBoostedTreesMaxBins{   def main(args: Array[String]) {     val data = GradientBoostedTreesUtil.getTrainTestData()     val train_data = data._1     val test_data = data._2     val maxBins_param = Array(10,16,32,64)     val iteration = 10     val maxDepth = 3     val i = 0     val resultsMap =      new scala.collection.mutable.HashMap[String, String]     val dataset = new DefaultCategoryDataset()     for(i <- 0 until maxBins_param.length) {       val maxBin = maxBins_param(i)       val rmsle = GradientBoostedTreesUtil.evaluate(train_data,  test_data,iteration,maxDepth, ...