private void setupSubstrate()
{
if (isMulti)
{
substrate = new SkirmishSubstrate(25, 15, 25, HyperNEATParameters.substrateActivationFunction);
network = substrate.generateMultiGenomeModulus(seedGenome.Decode(null), 5).Decode(null);
}
else
{
substrate = new SkirmishSubstrate(5, 3, 5, HyperNEATParameters.substrateActivationFunction);
network = substrate.generateGenome(seedGenome.Decode(null)).Decode(null);
}
}
public void calcFitness(string genomeFile, int type)
{
XmlDocument doc = new XmlDocument();
doc.Load(genomeFile);
NeatGenome genome = XmlNeatGenomeReaderStatic.Read(doc);
INetworkEvaluator eval;
if (type == 0)
{
eval = new CTRNNNetworkEvaluator(4, 12, 12);
}
else if (type == 1)
{
eval = new SUPGNetworkEvaluator(4, 12, 12);
}
else
{
doClune = true;
eval = new CluneNetworkEvaluator(20, 20, 20);
}
var tempNet = genome.Decode(null);
MessageBox.Show(eval.threadSafeEvaluateNetwork(tempNet)[0].ToString());
}
public GenomeVisualizerForm(NeatGenome genome)
{
InitializeComponent();
updatePicture = true;
_currentGenome = genome;
_currentSelectedConnectionIndex = -1;
ShowGenomeNetwork(genome);
ShowGenomeConnections(genome);
List <PointF> inputN = new List <PointF>();
List <PointF> outputN = new List <PointF>();
outputN = substrate.getNeuronGroupByType(1);
inputN = substrate.getNeuronGroupByType(0);
EvolvableSubstrate se = new EvolvableSubstrate();
hiddenNeurons = new List <PointF>();
connections = new ConnectionGeneList();
se.generateSubstrate(inputN, outputN, (INetwork)genome.Decode(null),
HyperNEATParameters.initialDepth,
(float)HyperNEATParameters.varianceThreshold,
(float)HyperNEATParameters.bandingThreshold,
(int)HyperNEATParameters.ESIterations,
(float)HyperNEATParameters.divisionThreshold,
HyperNEATParameters.maximumDepth,
(uint)inputN.Count, (uint)outputN.Count, ref connections, ref hiddenNeurons);
ShowHyperCubeSlide(genome);
}
/// <summary>
/// Constructor that manually generates a NN controller (used mostly for debugging).
/// </summary>
/// <param name="x">X coordinate of initial position.</param>
/// <param name="y">Y coordinate of intial position.</param>
/// <param name="numSensors">Number of sensor components. Each agent has one sensor composed of at least 3 components.</param>
public NNControlledCreature(Texture2D bodyTexture, float x, float y, float headingInRadians, Simulator sim, bool drawSensorField_in, bool trackPlanting_in, int numSensors, bool freezeAfterPlanting_in)
: base(bodyTexture, x, y, headingInRadians, sim, drawSensorField_in, trackPlanting_in, numSensors)
{
PlanterSubstrate substrate = new PlanterSubstrate(308, 4, 108, new BipolarSigmoid());
NeatGenome genome = substrate.generateGenome();
Controller = genome.Decode(ActivationFunctionFactory.GetActivationFunction("BipolarSigmoid"));
freezeAfterPlanting = freezeAfterPlanting_in;
frozen = false;
}
private void ShowHyperCubeSlide(NeatGenome genome)
{
float x = (float)numericUpDownX.Value;
float y = (float)numericUpDownY.Value;
float starty = -1.0f;
float startx;
INetwork net = genome.Decode(null);
float[] coordinates = new float[4];
coordinates[0] = x;
coordinates[1] = y;
double output;
if (pictureBox1 != null)
{
return;
}
Graphics g = pictureBox1.CreateGraphics();
while (starty < 1.0f)
{
startx = -1.0f;
coordinates[3] = starty;
while (startx < 1.0f)
{
coordinates[2] = startx;
net.ClearSignals();
net.SetInputSignals(coordinates);
net.MultipleSteps(3);
output = net.GetOutputSignal(0);
//HyberCubeSlidePanel.
if (output < 0.0f)
{
brush.Color = Color.FromArgb((int)(output * -255.0f), 0, 0);
}
else
{
brush.Color = Color.FromArgb(0, 0, (int)(output * 255.0f));
}
g.FillRectangle(brush, new Rectangle((int)(startx * 100.0f) + 100, 100 - (int)(starty * 100.0f), 1, 1));
//Show origin
startx += 0.01f;
}
starty += 0.01f;
}
brush.Color = Color.Green;
g.FillEllipse(brush, new Rectangle((int)(x * 100.0f) + 100, 100 - (int)(y * 100.0f), 5, 5));
}
/// <summary>
/// If using NEAT, this function simply sets the AgentBrain genome variable as the brain. Otherwise, the genome is actually used as a genome and is decoded into the appropriate brain structure.
/// </summary>
private void createBrains()
{
if (Genome != null)
{
if (NeatBrain)
{
Brain = Genome;
}
else
if (Homogeneous)
{
ANN = SubstrateDescription.generateHomogeneousGenome(Genome, NormalizeANNWeights, this.AdaptableANN, this.ModulatoryANN, EvolveSubstrate);
Brains = new List <INetwork>();
for (int i = 0; i < NumRobots; i++)
{
INetwork b = ANN.Decode(null);
Brains.Add(b);
}
}
else
{
if (MultipleBrains)
{
List <NeatGenome> genes = SubstrateDescription.generateGenomeStackSituationalPolicy(Genome, Convert.ToUInt32(NumRobots), NormalizeANNWeights, AdaptableANN, ModulatoryANN, 2, out ZCoordinates);
for (int j = 0; j < genes.Count; j++)
{
MultiBrains.Add(genes[j].Decode(null));
}
Brain = MultiBrains[0];
}
else if (Hive)
{
ANN = SubstrateDescription.generateHiveBrainGenomeStack(Genome, Convert.ToUInt32(NumRobots), NormalizeANNWeights, AdaptableANN,
ModulatoryANN, out ZCoordinates, EvolveSubstrate, UseCTRNNs);
Brain = UseCTRNNs ? ANN.DecodeToCTRNN() : ANN.Decode(null);
}
else
{
ANN = SubstrateDescription.generateMultiGenomeStack(Genome, Convert.ToUInt32(NumRobots), NormalizeANNWeights, AdaptableANN,
ModulatoryANN, out ZCoordinates, EvolveSubstrate);
Brain = ANN.Decode(null);
}
}
}
}
public double EvaluateNetwork(INetwork cppn)
{
lock (locker)
{
NeatGenome tempGenome = m_substrate.GenerateGenome(cppn);
INetwork tempSubsNet = tempGenome.Decode(null);
double fitness = 0.0;
Program.TheWorldSimulator.NetworksToEvaluate[m_agentId] = tempSubsNet;
Program.TheWorldSimulator.EventsNetworkReady[m_agentId].Set();
Program.TheWorldSimulator.EventsFitnessReady[m_agentId].WaitOne();
fitness = Program.TheWorldSimulator.Fitness;
//if (!SkirmishExperiment.multiple)
// fitness += doEvaluation(tempNet);
//else
// fitness += doEvaluationMulti(tempNet);
return(fitness);
}
}
private void createBrains()
{
if (genome != null)
{
if (homogenous)
{
ANN = substrateDescription.generateHomogeneousGenome(genome, normalizeANNWeights, this.adaptableANN, this.modulatoryANN, evolveSubstrate);
brains = new List <INetwork>();
for (int i = 0; i < numRobots; i++)
{
INetwork b = ANN.Decode(null);
brains.Add(b);
}
}
else
{
if (multipleBrains) //Multiple brains with situational policies
{
List <NeatGenome> genes = substrateDescription.generateGenomeStackSituationalPolicy(genome, Convert.ToUInt32(numRobots), normalizeANNWeights, adaptableANN, modulatoryANN, 2, out zcoordinates);
for (int j = 0; j < genes.Count; j++)
{
multiBrains.Add(genes[j].Decode(null));
}
brain = multiBrains[0];
}
else
{
ANN = substrateDescription.generateMultiGenomeStack(genome, Convert.ToUInt32(numRobots), normalizeANNWeights, adaptableANN,
modulatoryANN, out zcoordinates, evolveSubstrate);
brain = ANN.Decode(null);
}
}
}
}
private void createBrains()
{
if (genome != null)
{
//Schrum: debugging
//Console.WriteLine("genome != null");
if (homogenous)
{
//Schrum: debugging
//Console.WriteLine("homogenous");
ANN = substrateDescription.generateHomogeneousGenome(genome, normalizeANNWeights, this.adaptableANN, this.modulatoryANN, evolveSubstrate);
brains = new List <INetwork>();
for (int i = 0; i < numRobots; i++)
{
INetwork b = ANN.Decode(null);
brains.Add(b);
}
}
else
{
//Schrum: debugging
//Console.WriteLine("!homogenous");
if (multipleBrains) //Multiple brains with situational policies
{
//Schrum: debugging
//Console.WriteLine("multipleBrains");
// Schrum: The original code hard codes "2" as the nubmer of brains for any multi brain scenario TODO
//List<NeatGenome> genes = substrateDescription.generateGenomeStackSituationalPolicy(genome, Convert.ToUInt32(numRobots), normalizeANNWeights, adaptableANN, modulatoryANN, 2, out zcoordinates);
List <NeatGenome> genes = substrateDescription.generateGenomeStackSituationalPolicy(genome, Convert.ToUInt32(numRobots), normalizeANNWeights, adaptableANN, modulatoryANN, numBrains, out zcoordinates, forcedSituationalPolicyGeometry);
// Schrum: Debugging module deletion
/**
* if(genes.Count == 0)
* {
* Console.WriteLine("InputNeuronCount :" + genome.InputNeuronCount);
* Console.WriteLine("NumLinks :" + genome.NumLinks);
* Console.WriteLine("NumOutputModules :" + genome.NumOutputModules);
* Console.WriteLine("OutputNeuronCount:" + genome.OutputNeuronCount);
* Console.WriteLine("OutputsPerPolicy :" + genome.OutputsPerPolicy);
* Console.WriteLine("TotalNeuronCount :" + genome.TotalNeuronCount);
*
* throw new Exception("How is genes.Count 0!");
* }
**/
for (int j = 0; j < genes.Count; j++)
{
multiBrains.Add(genes[j].Decode(null));
}
// Schrum: for debugging
/*
* for (int j = 0; j < multiBrains.Count; j++)
* {
* if (multiBrains[j] == null)
* {
* Console.WriteLine(j +": Null brain!");
* }
* else
* {
* Console.WriteLine(j + ":" + multiBrains[j]);
* Console.WriteLine(j + ":CurrentOutputModule:" + multiBrains[j].CurrentOutputModule);
* Console.WriteLine(j + ":InputNeuronCount:" + multiBrains[j].InputNeuronCount);
* Console.WriteLine(j + ":OutputNeuronCount:" + multiBrains[j].OutputNeuronCount);
* Console.WriteLine(j + ":OutputsPerPolicy:" + multiBrains[j].OutputsPerPolicy);
* Console.WriteLine(j + ":TotalNeuronCount:" + multiBrains[j].TotalNeuronCount);
* Console.WriteLine(j + ":NumOutputModules:" + multiBrains[j].NumOutputModules);
* }
* }
*/
brain = multiBrains[0];
}
else
{
//Schrum: debugging
//Console.WriteLine("!multipleBrains");
ANN = substrateDescription.generateMultiGenomeStack(genome, Convert.ToUInt32(numRobots), normalizeANNWeights, adaptableANN,
modulatoryANN, out zcoordinates, evolveSubstrate);
brain = ANN.Decode(null);
}
}
}
}
public void showMovie(string genomeFile, int type)
{
if (true) // set to false to use hardcoded output values from a file
{
XmlDocument doc = new XmlDocument();
doc.Load(genomeFile);
NeatGenome genome = XmlNeatGenomeReaderStatic.Read(doc);
INetwork tempNet = null;
INetwork cppn = null;
NeatGenome tempGenome = null;
Substrate substrate;
if (type == 0)
{
substrate = new CTRNNSubstrate(4, 12, 8, HyperNEATParameters.substrateActivationFunction);
}
else if (type == 1)
{
substrate = new SUPGSubstrate(4, 12, 12, HyperNEATParameters.substrateActivationFunction);
}
else
{
doClune = true;
substrate = new CluneSubstrate(20, 20, 20, HyperNEATParameters.substrateActivationFunction);
}
cppn = genome.Decode(null);
tempGenome = substrate.generateGenome(cppn);
tempNet = tempGenome.Decode(null);
Controller controller;
if (type == 0)
{
controller = new Controller(tempNet);
}
else if (type == 1)
{
controller = new Controller(tempNet, true, tempGenome, cppn, ((SUPGSubstrate)substrate).getSUPGMap());
}
else
{
controller = new Controller(tempNet);
}
using (var domain = new Domain())
{
domain.Initialize(controller);
domain.RunDraw();
}
}
else
{
using (var domain = new Domain())
{
domain.Initialize();
domain.RunDraw();
}
}
}
private void ShowHyperCubeSlide(NeatGenome genome)
{
float posx = (float)numericUpDownX.Value;
float posy = (float)numericUpDownY.Value;
float x = (float)numericUpDownX.Value;
float y = (float)numericUpDownY.Value;
float starty = -1.0f;
float startx;
INetwork net = genome.Decode(null);
float[] coordinates = new float[4];
coordinates[0] = x;
coordinates[1] = y;
double output;
// if (pictureBox1 != null) return;
Graphics g = pictureBox1.CreateGraphics();
while (starty < 1.0f)
{
startx = -1.0f;
coordinates[3] = starty;
while (startx < 1.0f)
{
coordinates[2] = startx;
net.ClearSignals();
net.SetInputSignals(coordinates);
((ModularNetwork)net).RecursiveActivation();
output = net.GetOutputSignal(0);
//HyberCubeSlidePanel.
if (output < 0.0f)
{
brush.Color = Color.FromArgb((int)(output * -255.0f), 0, 0);
}
else
{
brush.Color = Color.FromArgb(0, 0, (int)(output * 255.0f));
}
g.FillRectangle(brush, new Rectangle((int)(startx * 100.0f) + 100, 100 - (int)(starty * 100.0f), 1, 1));
//Show origin
startx += 0.01f;
}
starty += 0.01f;
}
brush.Color = Color.Green;
//g.FillEllipse(brush, new Rectangle((int)(x * 100.0f) +100, 100-(int)(y * 100.0f), 5, 5));
brush.Color = Color.White;
float tolerance = (float)numericUpDownTolerance.Value;
if (connections != null)
{
foreach (ConnectionGene cg in connections)
{
if (Math.Abs(cg.coordinates[0] - posx) < tolerance && (Math.Abs(cg.coordinates[1] - posy) < tolerance))
{
g.FillEllipse(brush, new Rectangle((int)(cg.coordinates[2] * 100.0f) + 100, 100 - (int)(cg.coordinates[3] * 100.0f), 2, 2));
}
}
}
// redrawSlide = false;
g.DrawEllipse(blackPen, new Rectangle((int)((float)numericUpDownX.Value * 100.0f) + 100 - 4, 100 - (int)((float)numericUpDownY.Value * 100.0f) - 4, 8, 8));//new Rectangle((int)(mx-4), (int)(my-4), 8, 8));
brush.Color = Color.Green;
g.FillEllipse(brush, new Rectangle((int)(x * 100.0f) + 100, 100 - (int)(y * 100.0f), 5, 5));
}
public FieldSubs3LQTable(RLClientBase client, int numTeammates, int myUnum)
: base(client, numTeammates, myUnum, false)
{
if (Program.LoadingGenome)
{
try
{
var doc = new XmlDocument();
doc.Load(Program.LoadedGenomePath);
NeatGenome readCPPN = XmlNeatGenomeReaderStatic.Read(doc);
var substrate = new MultiLayerSandwichSubstrate(Rows, Cols + 2, NeatExpParams.SubstrateLayers,
NeatExpParams.AddBiasToSubstrate, HyperNEATParameters.substrateActivationFunction);
SharpNeatLib.NeatGenome.NeatGenome tempGenome = substrate.GenerateGenome(readCPPN.Decode(HyperNEATParameters.substrateActivationFunction));
m_loadedNetwork = tempGenome.Decode(HyperNEATParameters.substrateActivationFunction);
m_loadingBehaviour = true;
}
catch (Exception ex)
{
Console.WriteLine(ex.ToString());
m_loadingBehaviour = false;
}
}
}