mgo.test.TestNEATXOR.scala Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of mgo_3 Show documentation
mgo
The newest version!
///*
// * Copyright (C) 03/07/2015 Guillaume Chérel
// *
// * This program is free software: you can redistribute it and/or modify
// * it under the terms of the GNU General Public License as published by
// * the Free Software Foundation, either version 3 of the License, or
// * (at your option) any later version.
// *
// * This program is distributed in the hope that it will be useful,
// * but WITHOUT ANY WARRANTY; without even the implied warranty of
// * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// * GNU General Public License for more details.
// *
// * You should have received a copy of the GNU General Public License
// * along with this program.  If not, see .
// */
//
//package mgo.test
//
//import java.io.{ FileWriter, PrintWriter, File }
//import java.util.Locale
//
//import mgo.{ Individual, PopulationElement, Population }
//import mgo.evolution.algorithm.NEAT
//import mgo.evolution.breed.{ NEATSpeciesFitnessSharing, NEATSpeciesRescaledFitnessSharing, NEATFeedforwardTopology }
//import mgo.evolution.genome.{ NEATMinimalGenomeUnconnected, NEATMinimalGenomeConnectedIO }
//import mgo.termination.ConditionalTermination
//import mgo.tools.neuralnetwork.{ ActivationFunction, Feedforward, NeuralNetwork }
//
//import scala.collection.immutable.Map
//import scala.math._
//import scala.util.Random
//
//object TestNEATXOR {
//
//  implicit val rng = new Random
//
//  val outputDir = "/tmp/NEATXOR/output/"
//  val outputFile = outputDir ++ "singlerun.csv"
//  def outputBestNetDir = outputDir ++ "singleRunBestNet/"
//  def outputBestNetExt = ".gv"
//
//  def writeToFile(filename: String, string: String) {
//    val pw = new FileWriter(filename)
//    try {
//      pw.write(string)
//    } finally {
//      pw.close()
//    }
//  }
//
//  def appendToFile(filename: String, string: String) {
//    val pw = new FileWriter(filename, true)
//    try {
//      pw.write(string)
//    } finally {
//      pw.close()
//    }
//  }
//
//  def main(args: Array[String]): Unit = {
//
//    val xorneat = new XORNEAT {}
//
//    println("Creating " ++ outputDir)
//    new File(outputDir).mkdirs()
//    println("Creating " ++ outputBestNetDir)
//    new File(outputBestNetDir).mkdirs()
//
//    // delete previously generated graph files
//    new File(outputBestNetDir).listFiles().foreach { _.delete() }
//
//    val header = s"Generation\tPopsize\tNumSpecies\tcompatibility threshold\tbest\tavg\tworse\tnodes\tedges\t[(species,fitness,size)]\n"
//    println(header)
//    writeToFile(outputFile, header)
//
//    xorneat.evolve.untilConverged { s =>
//      val fitnesses: Vector[Double] = if (s.population.isEmpty) Vector(0.0) else s.population.map { elt => elt.fitness }.toVector
//      val bestfitness = fitnesses.max
//      val worsefitness = fitnesses.min
//      val averagefitness = fitnesses.sum / fitnesses.length
//      val compatibilitythreshold = s.archive.speciesCompatibilityThreshold.head
//
//      val bestNet = xorneat.createNet(s.population.maxBy { elt => elt.fitness }.phenotype)
//      val nodes = bestNet.network.nodes.length
//      val edges = bestNet.network.iteredges.length
//
//      val line = s"${s.generation}\t${s.population.content.length}\t${s.archive.indexOfSpecies.size}\t$compatibilitythreshold\t$bestfitness\t$averagefitness\t$worsefitness\t$nodes\t$edges\t${xorneat.speciesFitnessesOffsprings(s.population)}\n"
//      println(line)
//      appendToFile(outputFile, line)
//
//      writeToFile(outputBestNetDir ++ "%06d".format(s.generation) ++ outputBestNetExt, xorneat.dotString(bestNet))
//    }
//  }
//}
//
//object TestNEATXORReplications {
//
//  implicit val rng = new Random
//
//  val replications = 100
//
//  val outputDir = "/tmp/NEATXOR/output"
//  val outputFile = outputDir ++ "replications.csv"
//  val outputBestDir = outputDir ++ "replicationsBestNets/"
//  val outputBestExt = ".gv"
//
//  def main(args: Array[String]): Unit = {
//
//    val xorneat = new XORNEAT {}
//
//    new File(outputDir).mkdirs()
//    new File(outputBestDir).mkdirs()
//
//    // delete previously generated gv files
//    new File(outputBestDir).listFiles().foreach { _.delete() }
//
//    val fileoutput = new PrintWriter(outputFile)
//
//    try {
//      val header = s"Generation\tPopsize\tNumSpecies\tbest\tavg\tworse\tnodes\tedges\t[(species,fitness,size)]"
//      println(header)
//      fileoutput.println(header)
//
//      (1 to replications).foreach { repli =>
//        val finalstate = xorneat.evolve.untilConverged { s => }
//        val fitnesses: Vector[Double] = if (finalstate.population.isEmpty) Vector(0.0) else finalstate.population.map { elt => elt.fitness }.toVector
//        val bestfitness = fitnesses.max
//        val worsefitness = fitnesses.min
//        val averagefitness = fitnesses.sum / fitnesses.length
//        val speciesInfo = xorneat.speciesFitnessesOffsprings(finalstate.population)
//
//        val bestNet = xorneat.createNet(finalstate.population.maxBy { elt => elt.fitness }.phenotype)
//        val nodes = bestNet.network.nodes.length
//        val edges = bestNet.network.iteredges.length
//
//        val line = s"${finalstate.generation}\t${finalstate.population.content.length}\t${finalstate.archive.indexOfSpecies.size}\t$bestfitness\t$averagefitness\t$worsefitness\t$nodes\t$edges\t$speciesInfo}"
//        println(line)
//        fileoutput.println(line)
//
//        val filebestnetoutput = new PrintWriter(outputBestDir ++ "%06d".format(repli) ++ outputBestExt)
//        try {
//          filebestnetoutput.println(xorneat.dotString(bestNet))
//        } finally {
//          filebestnetoutput.close()
//        }
//
//      }
//    } finally {
//      fileoutput.close()
//    }
//  }
//}
//
//trait XORNEAT extends NEAT with NEATMinimalGenomeConnectedIO with NEATFeedforwardTopology with ConditionalTermination with NEATSpeciesFitnessSharing {
//
//  def interSpeciesMatingProb: Double = 0.001
//
//  def mutationAddNodeProb: Double = 0.02
//  def mutationAddLinkProb: Double = 0.03
//  def mutationAddLinkBiasProb: Double = 0.0
//  def mutationWeightDriftTo0: Double = 0.1
//  def mutationWeightHardMin: Double = Double.NegativeInfinity
//  def mutationWeightHardMax: Double = Double.PositiveInfinity
//
//  def mutationWeightSigma: Double = 4
//  def mutationWeightProb0: Double = 0.0
//
//  def mutationDisableProb: Double = 0.0
//  def mutationEnableProb: Double = 0.0
//
//  def genDistDisjointCoeff: Double = 1.0
//  def genDistExcessCoeff: Double = 1.0
//  def genDistWeightDiffCoeff: Double = 0.4
//
//  def numberSpeciesTarget: Int = 10
//
//  def speciesCompatibilityThreshold: Double = 3.0
//  def speciesCompatibilityMod: Double = 0.0
//  def speciesCompatibilityMin: Double = 0.0
//
//  def crossoverInheritDisabledProb: Double = 0.75
//
//  def proportionKeep: Double = 0.2
//
//  def speciesKeptIfStagnate: Int = 2
//  def stagnationTimeThreshold: Int = 0
//
//  def useSpeciesHint = true
//
//  val inputNodes = 2
//  val biasNodes = 1
//  val outputNodes = 1
//
//  val lambda = 150
//
//  type NN = NeuralNetwork[Double, Double, Double] with Feedforward[Double, Double]
//
//  val trainingSet: Seq[(Seq[Double], Seq[Double])] =
//    Vector(
//      (Vector(0.0, 0.0, 1.0), Vector(0.0)),
//      (Vector(0.0, 1.0, 1.0), Vector(1.0)),
//      (Vector(1.0, 0.0, 1.0), Vector(1.0)),
//      (Vector(1.0, 1.0, 1.0), Vector(0.0)))
//
//  def neuronInitValue: Double = 0.5
//
//  def createNet(
//    _nodes: IndexedSeq[Node],
//    _inputnodes: IndexedSeq[Int],
//    _outputnodes: IndexedSeq[Int],
//    _edges: Seq[(Int, Int, Double)])(implicit rng: Random): NN =
//    NeuralNetwork.feedforwardSparse[Double, Double, Double](
//      _nodes.map { _.level },
//      _inputnodes,
//      _outputnodes,
//      _edges,
//      ActivationFunction.tanh,
//      Vector.fill(_nodes.length)(neuronInitValue))
//
//  def evaluateNet(nn: NN)(implicit rng: Random): Double = {
//    val score = getScore(nn)
//    score.map { e: Seq[Double] => e.sum / e.length }.sum / score.length
//  }
//
//  /** returns expected and actual output difference for each output neuron and for each  */
//  def getScore(
//    nn: NeuralNetwork[Double, Double, Double] with Feedforward[Double, Double])(implicit rng: Random): Seq[Seq[Double]] = {
//    val shuffledset = rng.shuffle(trainingSet)
//    shuffledset.map {
//      case (input, output) =>
//        (nn.outputState(nn.query(input)) zip output).map { case (o, expected) => 1.0 - abs(expected - o) }
//    }
//  }
//
//  val maxSteps = 150
//  def terminated(population: Population[G, P, F])(implicit rng: Random): Boolean = {
//    val best: P = population.content.maxBy { elt: PopulationElement[G, P, F] => elt.fitness }.phenotype
//    val nn = createNet(best)
//    getScore(nn).forall(output => output.forall { 1 - _ < 0.45 })
//  }
//
//  val enLocale = new Locale("en")
//
//  def dotString(nn: NN): String =
//    nn.network.toDot(
//      "",
//      { n: Double => Vector(("level", n.toString)) },
//      { e: Double =>
//        Vector(
//          ("color",
//            if (e > 0.0) "red"
//            else if (e < 0.0) "blue"
//            else "black"),
//          ("label", "%.2f".formatLocal(enLocale, e)),
//          ("penwidth", abs(e).toString))
//      },
//      s"""rankdir=LR
//       |{rank=source ${nn.inputNeurons.mkString(" ")}}
//       |{rank=sink ${nn.outputNeurons.mkString(" ")}}""".stripMargin)
//
//  def speciesFitnessesOffsprings(
//    population: Population[G, P, F]): Seq[(Int, Double, Int)] = {
//    val indivsBySpecies: Map[Int, Seq[Individual[G, P, F]]] =
//      population.toIndividuals.groupBy { indiv => indiv.genome.species }
//    val speciesFitnesses: Seq[(Int, Double)] = indivsBySpecies.iterator.map {
//      case (sp, indivs) => (sp, indivs.map {
//        _.fitness
//      }.sum / indivs.size)
//    }.toSeq
//    val sumOfSpeciesFitnesses: Double = speciesFitnesses.map {
//      _._2
//    }.sum
//    val res: Seq[(Int, Double, Int)] = speciesFitnesses.map { case (sp, f) => (sp, f, round(f * lambda / sumOfSpeciesFitnesses).toInt) }
//    res.toVector
//  }
//}