/// <summary>
/// Main program.
/// </summary>
static void Main(string[] args)
{
// Get the selected fitness type.
IFitness myFitness = GetFitnessMethod();
// Genetic algorithm setup.
_ga = new GA(_crossoverRate, _mutationRate, 100, 10000000, _genomeSize);
// Get the target fitness for this method.
_targetFitness = myFitness.TargetFitness;
// Run the genetic algorithm and get the best brain.
string program = GAManager.Run(_ga, fitnessFunction, OnGeneration);
// Display the final program.
Console.WriteLine(program);
Console.WriteLine();
// Compile to executable.
BrainPlus.Compile(program, "output.exe", myFitness);
// Run the result for the user.
string result = myFitness.RunProgram(program);
Console.WriteLine(result);
Console.ReadKey();
}
public string Generate(IGeneticAlgorithm ga)
{
// Generate functions.
IFitness myFitness;
string originalTargetString = _targetParams.TargetString;
string program;
string appendCode = "";
// Split string into terms.
string[] parts = _targetParams.TargetString.Split(new char[] { ' ' });
// Build corpus of unique terms to generate functions.
Dictionary <string, string> terms = new Dictionary <string, string>();
foreach (string part in parts)
{
if (!string.IsNullOrEmpty(part))
{
terms[part] = part;
}
}
foreach (string term in terms.Values)
{
_targetParams.TargetString = term;
// Get the target fitness for this method.
myFitness = _getFitnessFunc();
_targetParams.TargetFitness = myFitness.TargetFitness;
// Run the genetic algorithm and get the best brain.
program = GAManager.Run(ga, _fitnessFunc, _generationFunc);
// Trim extraneous loop instructions from the end.
program = program.Replace("[]", "");
appendCode += program + "@";
// Reset the target fitness.
myFitness.ResetTargetFitness();
_bestStatus.Fitness = 0;
_bestStatus.TrueFitness = 0;
_bestStatus.Output = "";
_bestStatus.LastChangeDate = DateTime.Now;
_bestStatus.Program = "";
_bestStatus.Ticks = 0;
// Notify parent of progress.
_onStepComplete(appendCode, term);
}
// Restore target string.
_targetParams.TargetString = originalTargetString;
return(appendCode);
}
public string Generate(IGeneticAlgorithm ga)
{
// Generate functions.
IFitness myFitness;
string originalTargetString = _targetParams.TargetString;
string program;
string appendCode = "";
// Split string into terms of 3-characters.
string[] parts = SplitInParts(_targetParams.TargetString, _chunkSize).ToArray();
// Build corpus of unique terms to generate functions.
Dictionary <string, string> terms = new Dictionary <string, string>();
foreach (string part in parts)
{
terms[part] = part;
}
foreach (string term in terms.Values)
{
_targetParams.TargetString = term;
// Get the target fitness for this method.
myFitness = _getFitnessFunc();
_targetParams.TargetFitness = myFitness.TargetFitness;
// Run the genetic algorithm and get the best brain.
program = GAManager.Run(ga, _fitnessFunc, _generationFunc);
appendCode += program + "@";
// Reset the target fitness.
myFitness.ResetTargetFitness();
_bestStatus.Fitness = 0;
_bestStatus.TrueFitness = 0;
_bestStatus.Output = "";
_bestStatus.LastChangeDate = DateTime.Now;
_bestStatus.Program = "";
_bestStatus.Ticks = 0;
}
// Restore target string.
_targetParams.TargetString = originalTargetString;
return(appendCode);
}
/// <summary>
/// Main program.
/// </summary>
static void Main(string[] args)
{
// Genetic algorithm setup.
_ga = new GA(_crossoverRate, _mutationRate, 100, 10000000, _genomeSize);
if (_functionGenerator != null)
{
// Generate additional functions.
_appendCode += _functionGenerator.Generate(_ga);
}
// Generate main program. Instantiate the fitness method.
IFitness myFitness = GetFitness();
// Get the target fitness for this method.
_targetParams.TargetFitness = myFitness.TargetFitness;
// Run the genetic algorithm and get the best brain.
string program = GAManager.Run(_ga, fitnessFunction, OnGeneration);
// Append any functions.
if (!string.IsNullOrEmpty(_appendCode))
{
program += "@" + _appendCode;
}
// Display the final program.
Console.WriteLine(program);
Console.WriteLine();
// Compile to executable.
BrainPlus.Compile(program, "output.exe", myFitness);
// Run the result for the user.
string result = myFitness.RunProgram(program);
Console.WriteLine(result);
Console.ReadKey();
}
private void Awake()
{
Instance = this;
DontDestroyOnLoad(this);
}
private void GAComplete(GAManager gaManager)
{
RunComplete(gaManager.Results);
}
private void RunGA()
{
_projMgr.ModelRunning = true;
ResetGraphs();
ResetOutputFields();
double crossoverRate = Convert.ToDouble(txtCrossoverRate.Text);
double mutationRate = Convert.ToDouble(txtMutRate.Text);
int numGen = Convert.ToInt32(txtNumGen.Text);
int popSize = Convert.ToInt32(txtPopSize.Text);
FitnessCriteria fitnessCriteria = GetFitnessCriteria();
_maxIndVars = Convert.ToInt32(txtMaxVars.Text);
if (chkSeed.Checked)
{
int seed = Convert.ToInt32(txtSeed.Text);
RandomNumbers.SetRandomSeed(seed);
}
else
RandomNumbers.SetRandomSeed();
_maxVIF = Convert.ToInt32(txtMaxVIF.Text);
List<IIndividual> initPop = new List<IIndividual>(popSize);
for (int i = 0; i < popSize; i++)
{
initPop.Add(new MLRIndividual(_maxIndVars, _totVar, fitnessCriteria, _maxVIF, _decisionThreshold, _mandateThreshold));
//initPop.Add(new MLRIndividual(_maxIndVars, _totVar, fitnessCriteria, _maxVIF));
}
Population population = new Population(initPop);
population.CrossoverMethod = new MLROnePointCrossover(crossoverRate);
if (fitnessCriteria == FitnessCriteria.Akaike)
{
population.Comparer = new AscendSort();
AICSelector aicSelector = new AICSelector();
aicSelector.Comparer = population.Comparer;
population.Selector = aicSelector;
}
else if (fitnessCriteria == FitnessCriteria.AICC)
{
population.Comparer = new AscendSort();
AICCSelector aiccSelector = new AICCSelector();
aiccSelector.Comparer = population.Comparer;
population.Selector = aiccSelector;
}
else if (fitnessCriteria == FitnessCriteria.BIC)
{
population.Comparer = new AscendSort();
BICSelector bicSelector = new BICSelector();
bicSelector.Comparer = population.Comparer;
population.Selector = bicSelector;
}
else if (fitnessCriteria == FitnessCriteria.Press)
{
population.Comparer = new AscendSort();
PressSelector pressSelector = new PressSelector();
pressSelector.Comparer = population.Comparer;
population.Selector = pressSelector;
}
else if (fitnessCriteria == FitnessCriteria.AdjustedR2)
{
population.Comparer = new DescendSort();
AdjR2Selector adjR2Selector = new AdjR2Selector();
adjR2Selector.Comparer = population.Comparer;
population.Selector = adjR2Selector;
}
else if (fitnessCriteria == FitnessCriteria.R2)
{
population.Comparer = new DescendSort();
R2Selector r2Selector = new R2Selector();
r2Selector.Comparer = population.Comparer;
population.Selector = r2Selector;
}
else if (fitnessCriteria == FitnessCriteria.RMSE)
{
population.Comparer = new AscendSort();
RMSESelector rmseSelector = new RMSESelector();
rmseSelector.Comparer = population.Comparer;
population.Selector = rmseSelector;
}
else if (fitnessCriteria == FitnessCriteria.Sensitivity)
{
population.Comparer = new DescendSort();
SensitivitySelector sensitivitySelector = new SensitivitySelector();
sensitivitySelector.Comparer = population.Comparer;
population.Selector = sensitivitySelector;
}
else if (fitnessCriteria == FitnessCriteria.Specificity)
{
population.Comparer = new DescendSort();
SpecificitySelector specificitySelector = new SpecificitySelector();
specificitySelector.Comparer = population.Comparer;
population.Selector = specificitySelector;
}
else if (fitnessCriteria == FitnessCriteria.Accuracy)
{
population.Comparer = new DescendSort();
AccuracySelector accuracySelector = new AccuracySelector();
accuracySelector.Comparer = population.Comparer;
population.Selector = accuracySelector;
}
population.ChromosomeComparer = new CompareChromosomes();
population.Initialize();
population.Mutator = new MLRMutator(mutationRate, _totVar);
//GALib.GAManager ga = new GALib.GAManager();
_gaManager = new GALib.GAManager();
_gaManager.Init(population);
_gaManager.NumberOfGenerations = numGen;
_gaManager.GAProgress += new GAManager.GAProgressHandler(GAUpdate);
_gaManager.GAComplete += new GAManager.GACompleteHandler(GAComplete);
GAManager ga = new GAManager();
_runThread = new Thread(_gaManager.Run);
_runThread.Start();
//list = ga.Run(population);
}
