samplingLandMetrics.oms

// to execute from the command line: openmole -s samplingLandMetrics.oms

// output files
val agreOut = Val[File]
val f = Val[File]

// random number generator seed
val mySeed = Val[Int]

// model parameters
val ksi = Val[Double]
val y0 = Val[Double]
val sar = Val[Double]
val a = Val[Double]
val w = Val[Double]
val Tag = Val[Double]
val Tab = Val[Double]
val Tr = Val[Double]
val Td = Val[Double]
val d = Val[Double]
val mdc = Val[Double]
val ndc = Val[Double]

// simulation parameters
val T = Val[Double]
val dtp = Val[Double]
val L = Val[Double]
val d0 = Val[Double]
val a0 = Val[Double]
val dtsave = Val[Double]

// variable outputs
val meanES = Val[Double]
val giniES = Val[Double]
val avgDistance = Val[Double]
val stdDistance = Val[Double]

///////////////////////////////////////////////////////////////////////////////
// 1- MODEL PLUGIN AS AN EXECUTABLE FILE
///////////////////////////////////////////////////////////////////////////////

val scriptTask =
  SystemExecTask(
    command = Seq(
      "./alumss-exec ${T} ${dtp} ${L} ${a0} ${d0} ${ksi} ${y0} ${sar} ${a} ${w} ${Tag} ${Tab} ${Tr} ${Td} ${d} ${mdc} ${ndc} ${dtsave} ${mySeed} "),
    )set(
      (inputs, outputs) += (T,dtp,L,a0,d0,ksi,y0,sar,a,w,Tag,Tab,Tr,Td,d,mdc,ndc,dtsave,mySeed),

      // outputFiles is the path to the file containing the model outputs
      // outputFiles += ("DATA_AGRE_T_${T}_dtp_${dtp}_n_${L}_a0_${a0}_d0_${d0}_ksi_${ksi}_y0_${y0}_sar_${sar}_a_${a}_w_${w}_Tag_${Tag}_Tab_${Tab}_Tr_${Tr}_Td_${Td}_d_${d}_dtsave_${dtsave}_expid_${mySeed}.dat",agreOut),
      outputFiles += ("DATA_OUT",agreOut),
      // resources is the path to the executable that runs the simulations
      resources+="/home/AD.LSM.CNRS.FR/diego.bengochea-paz/ALUMSS/ALUMSS/alumss-exec",
      // resources+="/home/karpouzi/Research/Chapter2/gillespie-land-use/alumss-exec",

      // default parameter and initialization values
      T:=0.0,
      dtp:=0.1,
      L:=100,
      a0:=0.2,
      d0:=0.0,
      ksi:=1.2,
      y0:=0.2,
      sar:=0.25,
      a:=0.0,
      w:=0.0,
      Tag:=0.1,
      Tab:=50.0,
      Tr:=5.0,
      Td:=50.0,
      d:=1.0,
      mdc:=1.1,
      ndc:=1.1,
      dtsave:=4000.0,
      mySeed:=1111
    )

///////////////////////////////////////////////////////////////////////////////
// 2- ENVIRONMENT SETUP: USE 40 LOCAL CORES
///////////////////////////////////////////////////////////////////////////////

val env = LocalEnvironment(40)

///////////////////////////////////////////////////////////////////////////////
// 4 - DECLARATION OF NUMERICAL EXPERIMENTS
//////////////////////////////////////////////////////////////////////////////

// direct sampling for bifurcation diagram as a function of ndc

val readOutputFile =
    ScalaTask("""
    // read file as an array of lines and get last line
    val lastLineAgre = scala.io.Source.fromFile(agreOut).getLines.toList.last.mkString
    // split the string and store it in an array
    val valuesAgre = lastLineAgre.split(" ")

    val meanES = valuesAgre(0).toDouble
    val giniES = valuesAgre(1).toDouble
    val avgDistance = valuesAgre(2).toDouble
    val stdDistance = valuesAgre(3).toDouble
    """

    )set(
        inputs+=agreOut,
        (inputs, outputs)+=(a0,w,ndc),
        outputs+=(meanES,giniES,avgDistance,stdDistance)
    )

val runAndRead = MoleTask(scriptTask -- readOutputFile)

val replicationsLongTerm =
  Replication(
    evaluation = runAndRead,
    seed = mySeed,
    sample = 10,
    aggregation = Seq (
      w aggregate average,
      ndc aggregate average,
      a0 aggregate average,
      meanES aggregate average,
      giniES aggregate average,
      avgDistance aggregate average,
      stdDistance aggregate average)
  )

val longTermSampling =
DirectSampling(
  evaluation = replicationsLongTerm,
  sampling =
    (ndc in Seq(1.1,2.1,3.1,4.1,5.1,6.1,7.1,8.1,9.1,10.1))x
    (w in Seq(0.0,1.0,2.0,4.0))x
    (a0 in (0.1 to 0.9 by 0.05)),
)

longTermSampling on env hook (workDirectory / "experimentESMetricsNDC-130621.csv")