void ShowGenomeConnections(NeatGenome genome)
{
int savedIndex = listBoxConnections.SelectedIndex;
listBoxConnections.Items.Clear();
/* show information about each connection in the listbox */
foreach (ConnectionGene connection in genome.ConnectionGeneList)
{
NeuronGene sourceNeuron = genome.NeuronGeneList.GetNeuronById(connection.SourceNeuronId);
NeuronGene destinationNeuron = genome.NeuronGeneList.GetNeuronById(connection.TargetNeuronId);
string info = string.Format("{0}({1}) --> {2}({3}); weight:{4:F3})", connection.SourceNeuronId, sourceNeuron.NeuronType, connection.TargetNeuronId, destinationNeuron.NeuronType, connection.Weight);
/* add the info text and the conection object itself in the listitem */
ListItem item = new ListItem("", info, connection);
listBoxConnections.Items.Add(item);
}
/* try to update selected savedIndex and refresh the drawed network */
listBoxConnections.SelectedIndex = listBoxConnections.Items.Count > savedIndex ? savedIndex : -1;
// ShowNetworkFromGenome(genome);
}
private int getOffset(int leg, int cppnIterations, NeuronGene neuron)
{
int offset = 0;
var coordinates = new double[3];
coordinates[0] = neuron.XValue / compression;
cppn.ClearSignals();
cppn.SetInputSignals(coordinates);
cppn.MultipleSteps(cppnIterations);
float activation = cppn.GetOutputSignal(1);
offset = (int)Math.Ceiling((activation + 1) * wavelength / 2);
if (offset <= 0)
{
offset = 1;
}
if (offset >= wavelength)
{
offset = wavelength - 1;
}
return(offset);
}
internal NeuronGene GetRandomHiddenNeuronGene(NeuralNetworkGene networkGenes, int hiddenLayerIndex)
{
var neuronGene = new NeuronGene
{
Axon = new AxonGene
{
Weights = new List <double>(),
ActivationFunction = GetRandomActivationFunction().GetType()
},
Soma = new SomaGene
{
Bias = _weightInitializer.InitializeWeight(),
SummationFunction = GetRandomSummationFunction().GetType()
}
};
//update terminals for current neuron
LayerGene nextlayer = GetNextLayerGene(networkGenes, hiddenLayerIndex);
for (int i = 0; i < nextlayer.Neurons.Count; i++)
{
neuronGene.Axon.Weights.Add(_weightInitializer.InitializeWeight());
}
return(neuronGene);
}
private static void WriteNeuron(XmlElement xmlNeurons, NeuronGene neuronGene)
{
XmlElement xmlNeuron = XmlUtilities.AddElement(xmlNeurons, "neuron");
XmlUtilities.AddAttribute(xmlNeuron, "id", neuronGene.InnovationId.ToString());
XmlUtilities.AddAttribute(xmlNeuron, "type", XmlUtilities.GetNeuronTypeString(neuronGene.NeuronType));
XmlUtilities.AddAttribute(xmlNeuron, "activationFunction", neuronGene.ActivationFunction.FunctionId);
XmlUtilities.AddAttribute(xmlNeuron, "bias", neuronGene.Bias.ToString());
XmlUtilities.AddAttribute(xmlNeuron, "layer", neuronGene.Layer.ToString());
// newNode.attributes["layer"] = neuronGene.Layer;
}
//public static
public static NeatGenome Read(XmlElement xmlGenome)
{
int inputNeuronCount = 0;
int outputNeuronCount = 0;
uint id = uint.Parse(XmlUtilities.GetAttributeValue(xmlGenome, "id"));
//--- Read neuron genes into a list.
NeuronGeneList neuronGeneList = new NeuronGeneList();
XmlNodeList listNeuronGenes = xmlGenome.SelectNodes("neurons/neuron");
foreach (XmlElement xmlNeuronGene in listNeuronGenes)
{
NeuronGene neuronGene = ReadNeuronGene(xmlNeuronGene);
// Count the input and output neurons as we go.
switch (neuronGene.NeuronType)
{
case NeuronType.Input:
inputNeuronCount++;
break;
case NeuronType.Output:
outputNeuronCount++;
break;
}
neuronGeneList.Add(neuronGene);
}
//--- Read module genes into a list.
List <ModuleGene> moduleGeneList = new List <ModuleGene>();
XmlNodeList listModuleGenes = xmlGenome.SelectNodes("modules/module");
foreach (XmlElement xmlModuleGene in listModuleGenes)
{
moduleGeneList.Add(ReadModuleGene(xmlModuleGene));
}
//--- Read connection genes into a list.
ConnectionGeneList connectionGeneList = new ConnectionGeneList();
XmlNodeList listConnectionGenes = xmlGenome.SelectNodes("connections/connection");
foreach (XmlElement xmlConnectionGene in listConnectionGenes)
{
connectionGeneList.Add(ReadConnectionGene(xmlConnectionGene));
}
// Read behavior list
NeatGenome g = new NeatGenome(id, neuronGeneList, moduleGeneList, connectionGeneList, inputNeuronCount, outputNeuronCount);
XmlNode behaviorNode = xmlGenome.SelectSingleNode("behavior");
if (behaviorNode != null)
{
g.Behavior = ReadBehavior(behaviorNode); //TODO bug is here
g.Behavior.objectives = new double[6];
g.objectives = new double[6];
}
/*
* XmlNode behaviorNode = xmlGenome.SelectSingleNode("behavior");
* if (behaviorNode != null)
* {
* g.Behavior = ReadBehavior(behaviorNode.SelectSingleNode("list")); //TODO bug is here
* g.Behavior.objectives = new double[6];
* g.objectives = new double[6];
* }
*/
return(g);
}
private string getState(NeuronGene neuronGene)
{
return(((IMarkovActivationFunction)(_fnLib.GetFunction(neuronGene.ActivationFnId))).State);
}
internal INeuron CreateNeuronFromGene(NeuronGene neuronGene, Dictionary<int, Dictionary<int, IList<Synapse>>> mapping, int layerIndex, int neuronIndex)
{
var dendrites = (layerIndex > 0) ? getDendritesForSoma(layerIndex, neuronIndex, mapping) : mapping[layerIndex][neuronIndex];
var soma = _somaFactory.Create(dendrites, neuronGene.Soma.Bias, neuronGene.Soma.SummationFunction);
var terminals = mapping[layerIndex + 1][neuronIndex];
var axon = _axonFactory.Create(terminals, neuronGene.Axon.ActivationFunction);
return Neuron.GetInstance(soma, axon);
}
internal ConnectionNeurons(NeuronGene sourceNeuronGene, NeuronGene targetNeuronGene)
{
Source = sourceNeuronGene;
Target = targetNeuronGene;
}
internal NeuronGene TryMutateNeuron(NeuronGene gene, double mutateChance, out bool didMutate)
{
didMutate = false;
NeuronGene toReturn = new NeuronGene
{
Axon = new AxonGene
{
Weights = new List <double>(),
ActivationFunction = gene.Axon.ActivationFunction
},
Soma = new SomaGene
{
SummationFunction = gene.Soma.SummationFunction
}
};
//weights
for (int j = 0; j < gene.Axon.Weights.Count; j++)
{
if (_config.MutateSynapseWeights && _random.NextDouble() <= mutateChance)
{
didMutate = true;
double val = _random.NextDouble();
if (_random.NextDouble() < 0.5)
{
// 50% chance of being negative, being between -1 and 1
val = 0 - val;
}
toReturn.Axon.Weights.Add(val);
}
else
{
toReturn.Axon.Weights.Add(gene.Axon.Weights[j]);
}
}
//bias
if (_config.MutateSomaBiasFunction && _random.NextDouble() <= mutateChance)
{
didMutate = true;
double val = _random.NextDouble();
if (_random.NextDouble() < 0.5)
{
// 50% chance of being negative, being between -1 and 1
val = 0 - val;
}
toReturn.Soma.Bias = val;
}
else
{
toReturn.Soma.Bias = gene.Soma.Bias;
}
//activation
if (_config.MutateAxonActivationFunction && _random.NextDouble() <= mutateChance)
{
didMutate = true;
toReturn.Axon.ActivationFunction = GetRandomActivationFunction().GetType();
}
else
{
toReturn.Axon.ActivationFunction = gene.Axon.ActivationFunction;
}
//summation
if (_config.MutateSomaSummationFunction && _random.NextDouble() <= mutateChance)
{
didMutate = true;
toReturn.Soma.SummationFunction = GetRandomSummationFunction().GetType();
}
else
{
toReturn.Soma.SummationFunction = gene.Soma.SummationFunction;
}
return(gene);
}
public static NeatGenome Read(XmlElement xmlGenome)
{
int inputNeuronCount = 0;
int outputNeuronCount = 0;
uint id = uint.Parse(XmlUtilities.GetAttributeValue(xmlGenome, "id"));
//--- Read neuron genes into a list.
NeuronGeneList neuronGeneList = new NeuronGeneList();
XmlNodeList listNeuronGenes = xmlGenome.SelectNodes("neurons/neuron");
foreach (XmlElement xmlNeuronGene in listNeuronGenes)
{
NeuronGene neuronGene = ReadNeuronGene(xmlNeuronGene);
// Count the input and output neurons as we go.
switch (neuronGene.NeuronType)
{
case NeuronType.Input:
inputNeuronCount++;
break;
case NeuronType.Output:
outputNeuronCount++;
break;
}
neuronGeneList.Add(neuronGene);
}
//--- Read module genes into a list.
List <ModuleGene> moduleGeneList = new List <ModuleGene>();
XmlNodeList listModuleGenes = xmlGenome.SelectNodes("modules/module");
foreach (XmlElement xmlModuleGene in listModuleGenes)
{
moduleGeneList.Add(ReadModuleGene(xmlModuleGene));
}
//--- Read connection genes into a list.
ConnectionGeneList connectionGeneList = new ConnectionGeneList();
XmlNodeList listConnectionGenes = xmlGenome.SelectNodes("connections/connection");
foreach (XmlElement xmlConnectionGene in listConnectionGenes)
{
connectionGeneList.Add(ReadConnectionGene(xmlConnectionGene));
}
//return new NeatGenome(id, neuronGeneList, connectionGeneList, inputNeuronCount, outputNeuronCount);
NeatGenome g = new NeatGenome(id, neuronGeneList, moduleGeneList, connectionGeneList, inputNeuronCount, outputNeuronCount);
g.Behavior = ReadBehavior(xmlGenome.SelectSingleNode("behavior"));
g.Behavior.objectives = new double[6];
g.objectives = new double[6];
// JUSTIN: Read grid/trajectory info
g.GridCoords = ReadGrid(xmlGenome.SelectSingleNode("grid"));
g.Behavior.trajectory = ReadTrajectory(xmlGenome.SelectSingleNode("trajectory"));
return(g);
}
internal NeuronGene TryMutateNeuron(NeuronGene gene, double mutateChance, out bool didMutate)
{
didMutate = false;
NeuronGene toReturn = new NeuronGene
{
Axon = new AxonGene
{
Weights = new List<double>(),
ActivationFunction = gene.Axon.ActivationFunction
},
Soma = new SomaGene
{
SummationFunction = gene.Soma.SummationFunction
}
};
//weights
for (int j = 0; j < gene.Axon.Weights.Count; j++)
{
if (_config.MutateSynapseWeights && _random.NextDouble() <= mutateChance)
{
didMutate = true;
double val = _random.NextDouble();
if (_random.NextDouble() < 0.5)
{
// 50% chance of being negative, being between -1 and 1
val = 0 - val;
}
toReturn.Axon.Weights.Add(val);
}
else
{
toReturn.Axon.Weights.Add(gene.Axon.Weights[j]);
}
}
//bias
if (_config.MutateSomaBiasFunction && _random.NextDouble() <= mutateChance)
{
didMutate = true;
double val = _random.NextDouble();
if (_random.NextDouble() < 0.5)
{
// 50% chance of being negative, being between -1 and 1
val = 0 - val;
}
toReturn.Soma.Bias = val;
}
else
{
toReturn.Soma.Bias = gene.Soma.Bias;
}
//activation
if (_config.MutateAxonActivationFunction && _random.NextDouble() <= mutateChance)
{
didMutate = true;
toReturn.Axon.ActivationFunction = GetRandomActivationFunction().GetType();
}
else
{
toReturn.Axon.ActivationFunction = gene.Axon.ActivationFunction;
}
//summation
if (_config.MutateSomaSummationFunction && _random.NextDouble() <= mutateChance)
{
didMutate = true;
toReturn.Soma.SummationFunction = GetRandomSummationFunction().GetType();
}
else
{
toReturn.Soma.SummationFunction = gene.Soma.SummationFunction;
}
return gene;
}
internal NeuronGene GetRandomHiddenNeuronGene(NeuralNetworkGene networkGenes, int hiddenLayerIndex)
{
var neuronGene = new NeuronGene
{
Axon = new AxonGene
{
Weights = new List<double>(),
ActivationFunction = GetRandomActivationFunction().GetType()
},
Soma = new SomaGene
{
Bias = _weightInitializer.InitializeWeight(),
SummationFunction = GetRandomSummationFunction().GetType()
}
};
//update terminals for current neuron
LayerGene nextlayer = GetNextLayerGene(networkGenes, hiddenLayerIndex);
for (int i = 0; i < nextlayer.Neurons.Count; i++)
{
neuronGene.Axon.Weights.Add(_weightInitializer.InitializeWeight());
}
return neuronGene;
}
void generateRandom(int numberOfInputs, int numberOfOutputs)
{
#region genome generation
// Generate random Genome
genome = new Genome();
//-- Generate Neuron part (include input and output)
IDictionary <int, NeuronGene> neuronGenes = new Dictionary <int, NeuronGene>();
//---- Outputs
for (int i = 0; i < numberOfOutputs; i++)
{
NeuronGene ng = new NeuronGene(outputLayer, (Neuron.Type)Random.Range(0, (int)Neuron.Type.amountOfTypes), i, false, true);
neuronGenes.Add(ng.innovation, ng);
}
//---- Inputs
for (int i = 0; i < numberOfInputs; i++)
{
NeuronGene ng = new NeuronGene(inputLayer, Neuron.Type.input, i + numberOfOutputs, true, false);
neuronGenes.Add(ng.innovation, ng);
}
//---- Generate hidden 0
int nRandoms = Random.Range(2, 5);
for (int i = 0; i < nRandoms; i++)
{
NeuronGene ng = new NeuronGene(0, (Neuron.Type)Random.Range(0, (int)Neuron.Type.amountOfTypes), neuronIA.getInnovation(), false, false);
neuronGenes.Add(ng.innovation, ng);
}
genome.neuronGenes = neuronGenes;
//-- Generate Connection part
IDictionary <int, ConnectionGene> connectionGenes = new Dictionary <int, ConnectionGene>();
List <int> hasInput = new List <int>();
List <int> hasOutput = new List <int>();
foreach (var n in genome.neuronGenes)
{
if (n.Value.layer != inputLayer)
{
if (!hasInput.Contains(n.Value.innovation))
{
int[] candidates = getInputCandidates(n.Value.layer);
//Debug.Log(candidates.Length);
int chosenCandidate = candidates[Random.Range(0, candidates.Length)];
ConnectionGene cg = new ConnectionGene(connectionIA.getInnovation(), chosenCandidate, n.Value.innovation, Random.Range(-1f, 1f));
connectionGenes.Add(cg.innovation, cg);
hasInput.Add(n.Value.innovation);
hasOutput.Add(chosenCandidate);
//Debug.Log("Added input");
}
}
if (n.Value.layer != outputLayer)
{
if (!hasOutput.Contains(n.Value.innovation))
{
int[] candidates = getOutputCandidates(n.Value.layer);
int chosenCandidate = candidates[Random.Range(0, candidates.Length)];
ConnectionGene cg = new ConnectionGene(connectionIA.getInnovation(), n.Value.innovation, chosenCandidate, Random.Range(-1f, 1f));
connectionGenes.Add(cg.innovation, cg);
hasOutput.Add(n.Value.innovation);
hasInput.Add(chosenCandidate);
//Debug.Log("Added output");
}
}
}
genome.connectionGenes = connectionGenes;
//printGenome();
#endregion
#region creation from genome
// Create neurons from Genome
inputs = new int[numberOfInputs];
outputs = new int[numberOfOutputs];
int ii = 0, io = 0;
foreach (var gn in genome.neuronGenes.Values)
{
Neuron neuron = new Neuron(gn.type, gn.layer, gn.innovation);
neurons[gn.innovation] = neuron;
if (neuron.layer == inputLayer)
{
inputs[ii] = neuron.innovation;
}
else if (neuron.layer == outputLayer)
{
outputs[io] = neuron.innovation;
}
}
//Debug.Log("inputs: " + inputs.Length);
// Create connections from genome
foreach (var gc in genome.connectionGenes.Values)
{
neurons[gc.toNeuronInnovation].inputs.Add(new NeuronInput(neurons[gc.fromNeuronInnovation], gc.weight));
}
#endregion
}
