static void Main(string[] args)
{
DataTable table = LoadData("dataset.txt");
TGPConfig config = new TGPConfig("TGPConfig.xml");
//config.CrossoverRate = 0.35;
//config.MicroMutationRate = 0.15;
//config.MacroMutationRate = 0.45;
//config.ReproductionRate = 0.05;
config.ElitismRatio = 0.1;
TGPPop <TGPSolution> pop = new TGPPop <TGPSolution>(config);
pop.PopulationReplacement = TGPPop <TGPSolution> .PopulationReplacementMode.TinyGP;
pop.OperatorSet.AddOperator(new TGPOperator_Plus());
pop.OperatorSet.AddOperator(new TGPOperator_Minus());
pop.OperatorSet.AddOperator(new TGPOperator_Division());
pop.OperatorSet.AddOperator(new TGPOperator_Multiplication());
pop.OperatorSet.AddOperator(new TGPOperator_Sin());
pop.OperatorSet.AddOperator(new TGPOperator_Cos());
pop.OperatorSet.AddIfgtOperator();
for (int i = 1; i < 10; ++i)
{
pop.ConstantSet.AddConstant(string.Format("C{0}", i), i);
}
pop.VariableSet.AddVariable("x");
pop.VariableSet.AddVariable("y");
pop.CreateFitnessCase += (index) =>
{
SpiralFitnessCase fitness_case = new SpiralFitnessCase();
fitness_case.X = double.Parse(table.Rows[index]["X"].ToString());
fitness_case.Y = double.Parse(table.Rows[index]["Y"].ToString());
fitness_case.Label = int.Parse(table.Rows[index]["Label"].ToString());
return(fitness_case);
};
pop.GetFitnessCaseCount += () =>
{
return(table.Rows.Count);
};
pop.EvaluateObjectiveForSolution += (fitness_cases, solution, objective_index) =>
{
double fitness = 0;
for (int i = 0; i < fitness_cases.Count; i++)
{
SpiralFitnessCase fitness_case = (SpiralFitnessCase)fitness_cases[i];
int correct_y = fitness_case.Label;
int computed_y = fitness_case.ComputedLabel;
fitness += (correct_y == computed_y) ? 0 : 1;
}
return(fitness);
};
pop.BreedInitialPopulation();
while (!pop.IsTerminated)
{
pop.Evolve();
Console.WriteLine("Spiral Classification Generation: {0}", pop.CurrentGeneration);
Console.WriteLine("Global Fitness: {0}\tCurrent Fitness: {1}", pop.GlobalBestProgram.Fitness, pop.FindFittestProgramInCurrentGeneration().Fitness);
Console.WriteLine("Global Best Solution:\n{0}", pop.GlobalBestProgram);
}
Console.WriteLine(pop.GlobalBestProgram.ToString());
}
static void Main(string[] args)
{
DataTable table = LoadData();
TGPConfig config = new TGPConfig("TGPConfig.xml");
TGPPop <TGPSolution> pop = new TGPPop <TGPSolution>(config);
pop.OperatorSet.AddOperator(new TGPOperator_Plus());
pop.OperatorSet.AddOperator(new TGPOperator_Minus());
pop.OperatorSet.AddOperator(new TGPOperator_Division());
pop.OperatorSet.AddOperator(new TGPOperator_Multiplication());
pop.OperatorSet.AddOperator(new TGPOperator_Power());
pop.OperatorSet.AddIfltOperator();
for (int i = 1; i < 10; ++i)
{
pop.ConstantSet.AddConstant(string.Format("C{0}", i), i);
}
pop.VariableSet.AddVariable("X1");
pop.VariableSet.AddVariable("X2");
pop.CreateFitnessCase += (index) =>
{
MexicanHatFitnessCase fitness_case = new MexicanHatFitnessCase();
fitness_case.X1 = double.Parse(table.Rows[index]["X1"].ToString());
fitness_case.X2 = double.Parse(table.Rows[index]["X2"].ToString());
fitness_case.Y = double.Parse(table.Rows[index]["Y"].ToString());
return(fitness_case);
};
pop.GetFitnessCaseCount += () =>
{
return(table.Rows.Count);
};
pop.EvaluateObjectiveForSolution += (fitness_cases, solution, objective_index) =>
{
double fitness = 0;
for (int i = 0; i < fitness_cases.Count; i++)
{
MexicanHatFitnessCase fitness_case = (MexicanHatFitnessCase)fitness_cases[i];
double correct_y = fitness_case.Y;
double computed_y = fitness_case.ComputedY;
fitness += (correct_y - computed_y) * (correct_y - computed_y);
}
return(fitness);
};
pop.BreedInitialPopulation();
while (!pop.IsTerminated)
{
pop.Evolve();
Console.WriteLine("Mexican Hat Symbolic Regression Generation: {0}", pop.CurrentGeneration);
Console.WriteLine("Global Fitness: {0}\tCurrent Fitness: {1}", pop.GlobalBestProgram.Fitness.ToString("0.000"), pop.FindFittestProgramInCurrentGeneration().Fitness.ToString("0.000"));
Console.WriteLine("Global Best Solution:\n{0}", pop.GlobalBestProgram);
}
Console.WriteLine(pop.GlobalBestProgram.ToString());
}
static void Main(string[] args)
{
DataTable table = LoadData();
TGPConfig config = new TGPConfig("TGPConfig.xml");
// The problem is to minimize the sum of errors between predicted and actual function
config.IsMaximization = false;
// As specified in Chapter 4 of "A Field Guide to Genetic Programming"
config.PopulationSize = 4;
config.ElitismRatio = 0; //non elitist
config.CrossoverRate = 0.5; // subtree crossover rate set to 0.5
config.MacroMutationRate = 0.25; // subtree mutation rate set to 0.25;
config.MicroMutationRate = 0.0; // point mutation rate set to 0.0
config.ReproductionRate = 0.25; // reproduction rate set to 0.25
config.NormalizeEvolutionRates();
//Question 1: Is the performance normal for the GP?
config.MaximumDepthForCreation = 2;
config.MaximumDepthForCrossover = 2; // no tree size limit by setting a very large max depth
config.MaximumDepthForMutation = 2;
TGPPop <TGPSolution> pop = new TGPPop <TGPSolution>(config);
pop.ReproductionSelectionInstruction = new SelectionInstruction_RouletteWheel <TGPSolution>(); //use roulette wheel selection
// Function Set = {+, -, %, *} where % is protected division that returns 1 if the denominator is 0
pop.OperatorSet.AddOperator(new TGPOperator_Plus());
pop.OperatorSet.AddOperator(new TGPOperator_Minus());
pop.OperatorSet.AddOperator(new TGPOperator_Division());
pop.OperatorSet.AddOperator(new TGPOperator_Multiplication());
// Terminal Set = {R, x}
pop.ConstantSet.AddConstant("R", DistributionModel.GetUniform() * 10.0 - 5.0);
pop.VariableSet.AddVariable("x");
pop.CreateFitnessCase += (index) =>
{
SymRegFitnessCase fitness_case = new SymRegFitnessCase();
fitness_case.X = double.Parse(table.Rows[index]["X"].ToString());
fitness_case.Y = double.Parse(table.Rows[index]["Y"].ToString());
return(fitness_case);
};
pop.GetFitnessCaseCount += () =>
{
return(table.Rows.Count);
};
pop.EvaluateObjectiveForSolution += (fitness_cases, solution, objective_index) =>
{
double sum_of_error = 0;
for (int i = 0; i < fitness_cases.Count; i++)
{
SymRegFitnessCase fitness_case = (SymRegFitnessCase)fitness_cases[i];
double correct_y = fitness_case.Y;
double computed_y = fitness_case.ComputedY;
sum_of_error += System.Math.Abs(correct_y - computed_y);
}
return(sum_of_error);
};
pop.BreedInitialPopulation();
double error = pop.GlobalBestProgram.ObjectiveValue;
while (error > 0.1)
{
pop.Evolve();
error = pop.GlobalBestProgram.ObjectiveValue;
Console.WriteLine("Symbolic Regression Generation: {0}", pop.CurrentGeneration);
Console.WriteLine("Minimum Error: {0}", error.ToString("0.000"));
Console.WriteLine("Global Best Solution:\n{0}", pop.GlobalBestProgram);
}
Console.WriteLine(pop.GlobalBestProgram.ToString());
}