public void CreateNewGeneration()
{
mGenerationCount++;
mCurrentGeneration.Sort();
mNewGeneration = new List <IndividualDNA>();
mReproducers = new List <IndividualDNA>();
//Select individuals with best fitness for reproduction
for (int i = 0; i < NumberOfReproducers; i++)
{
mReproducers.Add(mCurrentGeneration[i]);
}
mReproducers.Sort();
//Create Mother and Father sets of individuals
List <IndividualDNA> Mothers = new List <IndividualDNA>();
List <IndividualDNA> Fathers = new List <IndividualDNA>();
CreateNewGeneration1(Mothers, Fathers);
//Check for new Bestfit individual
mCurrentGeneration.Sort();
if (Math.Abs(mCurrentGeneration[0].Fitness) < Math.Abs(mBestFitness.Fitness))
{
mBestFitness = mCurrentGeneration[0];
}
SaveReproducers();
Save();
}
public string ToEquation(IndividualDNA individual)
{
string equation = String.Empty;
int exponent = 0;
string op = String.Empty;
foreach (double gene in individual.Genes)
{
if (gene >= 0)
{
op = " + ";
}
else
{
op = "";
}
if (equation.Equals(String.Empty))
{
op = "";
}
equation = equation + op + gene.ToString() + "x^" + exponent.ToString() + " ";
exponent++;
}
return(equation);
}
private void CreateNewGeneration1(List <IndividualDNA> Mothers, List <IndividualDNA> Fathers)
{
Mothers = mReproducers;
Fathers = mReproducers;
IndividualDNA child = new IndividualDNA();
//Add the top n to the new generation
mNewGeneration.Add(mCurrentGeneration[0]);
mNewGeneration.Add(mCurrentGeneration[1]);
//Perform cartesian product of Mothers and Fathers for reproduction
foreach (IndividualDNA mother in Mothers)
{
foreach (IndividualDNA father in Fathers)
{
if (!father.Equals(mother))
{
child = Reproduce(mother, father);
child.Generation = GenerationCount;
mNewGeneration.Add(child);
}
}
}
//if mNewGeneration is too small, fill it up with the
//best from mCurrentGeneration
int currentcount = 0;
while (mNewGeneration.Count < mPopulationSize)
{
mNewGeneration.Add(mCurrentGeneration[currentcount % 50]);
currentcount++;
}
//Calculate fitness of mNewGeneration
EvaluateIndividuals(mNewGeneration);
mNewGeneration.Sort();
//Make mNewGeneration the mCurrentGeneration
mCurrentGeneration.Clear();
for (int i = 0; i < mPopulationSize; i++)
{
if (!mCurrentGeneration.Contains(mNewGeneration[i]))
{
mCurrentGeneration.Add(mNewGeneration[i]);
}
}
currentcount = 0;
while (mCurrentGeneration.Count < mPopulationSize)
{
mCurrentGeneration.Add(mNewGeneration[currentcount % 50]);
currentcount++;
}
}
public IndividualDNA Crossover1(IndividualDNA _mother, IndividualDNA _father)
{
IndividualDNA child = new IndividualDNA(_mother.Terms);
child = _mother;
for (int i = 0; i < mCrossoverPoint; i++)
{
child.Genes[i] = _father.Genes[i];
}
return(child);
}
public int CompareTo(object obj)
{
int result = 1;
if (obj != null && obj is IndividualDNA)
{
IndividualDNA w = obj as IndividualDNA;
double comparex = Math.Abs(this.Fitness);
double comparey = Math.Abs(w.Fitness);
result = comparex.CompareTo(comparey);
}
return(result);
}
static public int Compare(IndividualDNA x, IndividualDNA y)
{
int result = 1;
double comparex = 0;
double comparey = 0;
if (x != null && y != null)
{
comparex = Math.Abs(x.Fitness);
comparey = Math.Abs(y.Fitness);
result = comparex.CompareTo(comparey);
}
return(result);
}
public IndividualDNA RandomMutation(IndividualDNA _individual)
{
IndividualDNA MutatedDNA = new IndividualDNA();
MutatedDNA = _individual;
Random rand = new Random(DateTime.Now.Millisecond);
Random rand2 = new Random(Convert.ToInt32(10000 * rand.Next(-100, 100)));
if (rand2.NextDouble() <= mChanceOfMutation)
{
MutatedDNA.Genes[rand2.Next(Terms / 2, Terms - 1)] = rand2.Next(-100, 100);
}
return(MutatedDNA);
}
public void CreateInitialRandomGeneration()
{
IndividualDNA individual;
for (int i = 0; i < mPopulationSize; i++)
{
individual = new IndividualDNA(mTerms, MinMaxFunctionValueY(FunctionValues));
individual.CreateInitialRandomGeneration(mCurrentGeneration.Count);
mCurrentGeneration.Add(individual);
}
EvaluateIndividuals(mCurrentGeneration);
mCurrentGeneration.Sort();
mBestFitness = mCurrentGeneration[0];
Save();
}
public IndividualDNA Crossover3(IndividualDNA _mother, IndividualDNA _father)
{
IndividualDNA child = new IndividualDNA(_mother.Terms);
for (int i = 0; i < Terms; i++)
{
if (i % 2 == 0)
{
child.Genes[i] = _mother.Genes[i];
}
else
{
child.Genes[i] = _father.Genes[i];
}
}
return(RandomMutation(child));
}
public void CreateInitialRandomGeneration()
{
IndividualDNA individual;
for (int i = 0; i < mPopulationSize; i++)
{
individual = new IndividualDNA(mTerms, MinMaxFunctionValueY(FunctionValues));
individual.CreateInitialRandomGeneration(mCurrentGeneration.Count);
mCurrentGeneration.Add(individual);
}
EvaluateIndividuals(mCurrentGeneration);
mCurrentGeneration.Sort();
mBestFitness = mCurrentGeneration[0];
Save();
}
public IndividualDNA Crossover1(IndividualDNA _mother, IndividualDNA _father)
{
IndividualDNA child = new IndividualDNA(_mother.Terms);
child = _mother;
for (int i = 0; i < mCrossoverPoint; i++)
{
child.Genes[i] = _father.Genes[i];
}
return child;
}
public IndividualDNA Crossover3(IndividualDNA _mother, IndividualDNA _father)
{
IndividualDNA child = new IndividualDNA(_mother.Terms);
for (int i = 0; i < Terms; i++)
{
if (i % 2 == 0)
{
child.Genes[i] = _mother.Genes[i];
}
else
{
child.Genes[i] = _father.Genes[i];
}
}
return RandomMutation(child);
}
public IndividualDNA Reproduce(IndividualDNA _mother, IndividualDNA _father)
{
return(Crossover3(_mother, _father));
}
public IndividualDNA RandomMutation(IndividualDNA _individual)
{
IndividualDNA MutatedDNA = new IndividualDNA();
MutatedDNA = _individual;
Random rand = new Random(DateTime.Now.Millisecond);
Random rand2 = new Random(Convert.ToInt32(10000 * rand.Next(-100, 100)));
if (rand2.NextDouble() <= mChanceOfMutation)
{
MutatedDNA.Genes[rand2.Next(Terms/2, Terms - 1)] = rand2.Next(-100, 100);
}
return MutatedDNA;
}
public IndividualDNA Reproduce(IndividualDNA _mother, IndividualDNA _father)
{
return Crossover3(_mother, _father);
}
public void CreateNewGeneration()
{
mGenerationCount++;
mCurrentGeneration.Sort();
mNewGeneration = new List<IndividualDNA>();
mReproducers = new List<IndividualDNA>();
//Select individuals with best fitness for reproduction
for (int i = 0; i < NumberOfReproducers; i++)
{
mReproducers.Add(mCurrentGeneration[i]);
}
mReproducers.Sort();
//Create Mother and Father sets of individuals
List<IndividualDNA> Mothers = new List<IndividualDNA>();
List<IndividualDNA> Fathers = new List<IndividualDNA>();
CreateNewGeneration1(Mothers, Fathers);
//Check for new Bestfit individual
mCurrentGeneration.Sort();
if (Math.Abs(mCurrentGeneration[0].Fitness) < Math.Abs(mBestFitness.Fitness)) mBestFitness = mCurrentGeneration[0];
SaveReproducers();
Save();
}
public IndividualDNAEnum(IndividualDNA[] list)
{
_DNA = list;
}
public static int Compare(IndividualDNA x, IndividualDNA y)
{
int result = 1;
double comparex = 0;
double comparey = 0;
if (x != null && y != null)
{
comparex = Math.Abs(x.Fitness);
comparey = Math.Abs(y.Fitness);
result = comparex.CompareTo(comparey);
}
return result;
}
private void CreateNewGeneration1(List<IndividualDNA> Mothers, List<IndividualDNA> Fathers)
{
Mothers = mReproducers;
Fathers = mReproducers;
IndividualDNA child = new IndividualDNA();
//Add the top n to the new generation
mNewGeneration.Add(mCurrentGeneration[0]);
mNewGeneration.Add(mCurrentGeneration[1]);
//Perform cartesian product of Mothers and Fathers for reproduction
foreach (IndividualDNA mother in Mothers)
{
foreach (IndividualDNA father in Fathers)
{
if (!father.Equals(mother))
{
child = Reproduce(mother, father);
child.Generation = GenerationCount;
mNewGeneration.Add(child);
}
}
}
//if mNewGeneration is too small, fill it up with the
//best from mCurrentGeneration
int currentcount = 0;
while (mNewGeneration.Count < mPopulationSize)
{
mNewGeneration.Add(mCurrentGeneration[currentcount%50]);
currentcount++;
}
//Calculate fitness of mNewGeneration
EvaluateIndividuals(mNewGeneration);
mNewGeneration.Sort();
//Make mNewGeneration the mCurrentGeneration
mCurrentGeneration.Clear();
for (int i = 0; i < mPopulationSize; i++)
{
if (!mCurrentGeneration.Contains(mNewGeneration[i]))
{
mCurrentGeneration.Add(mNewGeneration[i]);
}
}
currentcount = 0;
while (mCurrentGeneration.Count < mPopulationSize)
{
mCurrentGeneration.Add(mNewGeneration[currentcount%50]);
currentcount++;
}
}
public string ToEquation(IndividualDNA individual)
{
string equation = String.Empty;
int exponent = 0;
string op = String.Empty;
foreach (double gene in individual.Genes)
{
if (gene >= 0)
{
op = " + ";
}
else
{
op = "";
}
if (equation.Equals(String.Empty)) op = "";
equation = equation + op + gene.ToString() + "x^" + exponent.ToString() + " ";
exponent++;
}
return equation;
}
