LogisticClassifier_file.oms

import java.text.SimpleDateFormat
import java.util.Date

// generate basic binary data with
// the modified fire model

val forest_density = Val[Double]
val tree_resistance = Val[Double]
val replication = Val[Int]

val burnt = Val[Double]
val binaryburnt = Val[Double]

val simmodel = PythonTask(workDirectory / "fire.py") set (
    (inputs,outputs) += (forest_density,tree_resistance,replication),
    inputs += forest_density mapped "forest_density",
    inputs += tree_resistance mapped "tree_resistance",
    inputs += replication mapped "seed",
    outputs += burnt mapped "burnt",
    outputs += binaryburnt mapped "binary-burnt"
  )

val training = Val[File]
val validation = Val[File]

def datagen(outputfile: Val[File]) = {

  val aggregtofile = ScalaTask(s"""
    val ${outputfile.name} = newFile()
    ${outputfile.name}.content = (
      Seq("forest_density,tree_resistance,replication,burnt,binaryburnt")++
      ((Seq(forest_density.toSeq,tree_resistance.toSeq,replication.toSeq,burnt.toSeq,binaryburnt.toSeq).transpose).map(_.mkString(",")))).mkString("\\n")
   """
  ) set (
    inputs += (forest_density.toArray, resistance.toArray, replication.toArray, burnt.toArray, binaryburnt.toArray),
    outputs += outputfile
  )

  DirectSampling (
    evaluation = simmodel,
    sampling =
      ((forest_density in (0.1 to 0.9 by 0.1)) x
       (tree_resistance in (0.1 to 0.9 by 0.1)) x
       (replication in (UniformDistribution[Int](10000) take 5))),
    aggregation = Seq(forest_density, tree_resistance, replication, burnt, binaryburnt, traversed)

  ) -- aggregtofile

}

val errdensity = Val[Array[Double]]
val errresistance = Val[Array[Double]]
val score = Val[Double]

val sklearnclassifier =
  PythonTask(
    workDirectory / "logisticregression.py",
    libraries = Seq("pandas","numpy","sklearn")
  ) set (
    inputs += training mapped "data/training.csv",
    inputs += validation mapped "data/validation.csv",
    outputs += errdensity mapped "errdensity",
    outputs += errresistance mapped "errresistance",
    outputs += score mapped "score"
  )

val indics_hook = AppendToCSVFileHook(workDirectory / "exploration" / "classiferrors.csv")

EmptyTask() -- Seq(training, validation).map(datagen) -- (sklearnclassifier hook indics_hook)