public void MutateActivation(DenseRandomParams drp, int mutateRate)
{
if (random.Next(100) < mutateRate)
{
if (random.Next(100) < 10)
{
sameActivations = !sameActivations;
if (sameActivations)
{
int absorberIndex = random.Next(denseLayers.Count);
foreach (var t in denseLayers)
{
t.activationIndex = denseLayers[absorberIndex].activationIndex;
}
}
}
else
{
foreach (var t in denseLayers)
{
t.MutateActivation(drp, mutateRate);
}
}
}
}
} // конструктор для инициализации объекта DenseStructure по полям объекта ConvolutionalNetwork
public void MutateLayersNumb(DenseRandomParams drp, int mutateRate)
{
if (random.Next(100) < mutateRate)
{
denseLayersNumb = random.Next(1, drp.denseLayersNumbRange);
}
if (denseLayersNumb >= denseLayers.Count)
{
denseLayers.Capacity = denseLayersNumb;
int powIndex = Support.GetPow2(denseLayers[denseLayers.Count - 1].neurons);
while (denseLayers.Count != denseLayers.Capacity)
{
powIndex += random.Next(0, 2);
int neurons = (int)Math.Pow(2, powIndex);
denseLayers.Insert(0, new DenseLayer(drp, random, neurons));
if (sameActivations)
{
denseLayers[0].activationIndex = denseLayers[1].activationIndex;
}
}
}
else
{
denseLayers.RemoveRange(0, denseLayers.Capacity - denseLayersNumb); // перепроверить!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
denseLayers.Capacity = denseLayersNumb;
}
}
public void MutateDropouts(DenseRandomParams drp, int mutateRate)
{
if (random.Next(100) < mutateRate)
{
if (random.Next(100) < 10)
{
if (denseLayers[0].dropoutExist)
{
foreach (var t in denseLayers)
{
t.dropoutExist = false;
t.dropoutRate = 0;
}
}
else
{
foreach (var t in denseLayers)
{
t.dropoutExist = true;
t.dropoutRate = random.Next(10, drp.denseDropRange);
}
}
}
else
{
foreach (var t in denseLayers)
{
t.MutateDropout(drp, mutateRate);
}
}
}
}
public ConvolutionalChromosome(ConvolutionalNetwork cNet, ConvRandomParams crp, DenseRandomParams drp)
{
this.crp = crp;
this.drp = drp;
//добавить рандомные параметры
convPart = new ConvStructure(cNet.slidingWindows, cNet.convActivationIndexes, cNet.filters, cNet.convDropoutIndexes, cNet.convDropoutRates);
densePart = new DenseStructure(cNet.denseActivationIndexes, cNet.neurons, cNet.denseDropoutIndexes, cNet.denseDropoutRates);
} // конструктор для создания хромосомы из полей описательного класса ConvolutionalNetwork для последующего использования в генетических алгоритмах (например, для поиска лучшего варианта чем текущий)
} // псевдокопирующий конструктор
public void MutateActivation(DenseRandomParams drp, int mutateRate)
{
if (random.Next(100) < mutateRate)
{
activationIndex = random.Next(drp.denseActIndexesRange);
}
// поменять переменную sameActivations в другом классе(классе выше)
}
public DenseLayer(DenseRandomParams drp, Random random, int neurons)
{
this.random = random;
dropoutExist = random.Next(100) < 10 ? true : false;
if (dropoutExist)
{
dropoutRate = random.Next(10, drp.denseDropRange);
}
else
{
dropoutRate = 0;
}
activationIndex = random.Next(drp.denseActIndexesRange);
this.neurons = neurons;
}
public GeneticProgramm(NetworkRandomParams nrp, ConvRandomParams crp, DenseRandomParams drp,
int genEpochs, int[] selection, int populationSize, int mutateRate,
float accPriority, float paramPriority, int assessmentIndex, int percent)
{
_nrp = nrp;
_crp = crp;
_drp = drp;
_genEpochs = genEpochs;
_selection = selection;
_populationSize = populationSize;
_mutateRate = mutateRate;
_accPriority = accPriority;
_paramPriority = paramPriority;
_assessmentIndex = assessmentIndex;
_percent = percent;
}
public void MutateDense(DenseRandomParams drp, int mutateRate)
{
if (random.Next(100) < mutateRate)
{
if (random.Next(100) < 5)
{
densePart = new DenseStructure(nrp, drp, random);
}
else
{
densePart.MutateLayersNumb(drp, mutateRate);
densePart.MutateActivation(drp, mutateRate);
densePart.MutateNeurons(drp, mutateRate);
densePart.MutateDropouts(drp, mutateRate);
}
}
}
public ConvolutionalChromosome(GeneticProgramm gp, Random random)
{
_gp = gp;
this.random = random;
nrp = gp._nrp;
crp = gp._crp;
drp = gp._drp;
convPart = new ConvStructure(crp, random);
densePart = new DenseStructure(nrp, drp, random);
if (nrp.notRandomSpeed)
{
trainConstSpeed = nrp.trainConstSpeedRange[0];
}
else
{
trainConstSpeed = Convert.ToSingle(random.Next((int)(nrp.trainConstSpeedRange[0] * 1000), (int)(nrp.trainConstSpeedRange[1] * 1000))) / 1000;
}
optimizer = nrp.optimizers[random.Next(nrp.optimizers.Count)];
loss_function = nrp.loss_functions[random.Next(nrp.loss_functions.Count)];
}
public DenseStructure(NetworkRandomParams nrp, DenseRandomParams drp, Random random)
{
this.nrp = nrp;
this.drp = drp;
this.random = random;
sameActivations = random.Next(100) < 10 ? true : false;
int absorber = 0;
if (sameActivations)
{
absorber = random.Next(drp.denseActIndexesRange);
}
denseLayersNumb = random.Next(1, drp.denseLayersNumbRange);
denseLayers = new List <DenseLayer>(denseLayersNumb);
int powIndex = 0;
for (int i = 0; i < denseLayers.Capacity; i++)
{
int neurons = 0;
if (i == 0)
{
powIndex = random.Next(Support.GetPow2(nrp.networkOutputNumb), drp.firstDenseNeuronsRange);
}
else
{
powIndex += random.Next(0, 2);
}
neurons = (int)Math.Pow(2, powIndex);
denseLayers.Insert(0, new DenseLayer(drp, random, neurons));
if (sameActivations)
{
denseLayers[0].activationIndex = absorber;
}
}
}
public void MutateNeurons(DenseRandomParams drp, int mutateRate) // проверить
{
if (random.Next(100) < mutateRate)
{
int mutateIndex = random.Next(0, denseLayers.Count);
int powIndex = 0;
while (mutateIndex != -1)
{
if (mutateIndex != (denseLayers.Count - 1))
{
powIndex = Support.GetPow2(denseLayers[mutateIndex + 1].neurons);
}
else
{
powIndex = drp.firstDenseNeuronsRange;
}
powIndex += random.Next(0, 2);
denseLayers[mutateIndex].neurons = (int)Math.Pow(2, powIndex);
mutateIndex--;
}
}
}
public void MutateDropout(DenseRandomParams drp, int mutateRate)
{
if (random.Next(100) < mutateRate)
{
int mutationWay = random.Next(100);
if (mutationWay < 10)
{
dropoutExist = !dropoutExist;
if (dropoutExist)
{
dropoutRate = random.Next(10, drp.denseDropRange);
}
else
{
dropoutRate = 0;
}
}
else
{
dropoutRate = random.Next(10, drp.denseDropRange);
}
}
}
public GeneticProgramm CollectConv_NN_Params()
{
List <String> convActivations = new List <string>();
if (reluConvChb.Checked)
{
convActivations.Add("relu");
}
if (softmaxConvChb.Checked)
{
convActivations.Add("softmax");
}
if (lReluConvChb.Checked)
{
convActivations.Add("Leaky ReLU");
}
if (softsignConvChb.Checked)
{
convActivations.Add("softsign");
}
if (sigmoidConvChb.Checked)
{
convActivations.Add("sigmoid");
}
if (eluConvChb.Checked)
{
convActivations.Add("elu");
}
if (seluConvChb.Checked)
{
convActivations.Add("selu");
}
if (softplusConvChb.Checked)
{
convActivations.Add("softplus");
}
if (tanhConvChb.Checked)
{
convActivations.Add("tang");
}
if (PReLUConvChB.Checked)
{
convActivations.Add("PReLu");
}
if (TReLUConvChB.Checked)
{
convActivations.Add("TReLU");
}
List <String> denseActivations = new List <string>();
if (reluDenseChb.Checked)
{
denseActivations.Add("relu");
}
if (softmaxDenseChb.Checked)
{
denseActivations.Add("softmax");
}
if (lReluDenseChb.Checked)
{
denseActivations.Add("Leaky ReLU");
}
if (softsignDenseChb.Checked)
{
denseActivations.Add("softsign");
}
if (sigmoidDenseChb.Checked)
{
denseActivations.Add("sigmoid");
}
if (eluDenseChb.Checked)
{
denseActivations.Add("elu");
}
if (seluDenseChb.Checked)
{
denseActivations.Add("selu");
}
if (softplusDenseChb.Checked)
{
denseActivations.Add("softplus");
}
if (tanhDenseChb.Checked)
{
denseActivations.Add("tang");
}
if (PReLUDenseChB.Checked)
{
denseActivations.Add("PReLU");
}
if (ThReLUDenseChB.Checked)
{
denseActivations.Add("TReLU");
}
List <String> optimazers = new List <string>();
if (SGD_OptChB.Checked)
{
optimazers.Add("SGD");
}
List <String> lossFunctions = new List <string>();
if (categorical_crossentropyChB.Checked)
{
lossFunctions.Add("categorical_crossentropy");
}
List <int> callbacks_indexes = new List <int>();
if (modelCPChB.Checked)
{
callbacks_indexes.Add(0);
}
var nrp = new NetworkRandomParams(constSpeedChB.Checked,
new[] { float.Parse(minConstSpeedTB.Text, System.Globalization.CultureInfo.InvariantCulture),
float.Parse(maxConstSpeedTB.Text, System.Globalization.CultureInfo.InvariantCulture) },
datasetPathTB.Text, modelsPathTB.Text, labelsPathTB.Text, plotsPathTB.Text,
networkNameTB.Text, optimazers, lossFunctions, (int)learningEpochsNUD.Value, (int)batchSizeNUD.Value);
var crp = new ConvRandomParams((int)ConvLayersNumbNUD.Value,
(int)ConvFiltersNUD.Value,
convActivations.Count,
convActivations,
new[] { (int)slidingWindow1NUD.Value, (int)slidingWindow2NUD.Value },
convDropoutChB.Checked,
(int)convDropoutNUD.Value);
var drp = new DenseRandomParams((int)DenseLayersNumbNUD.Value,
(int)DenseNeuronsNUD.Value,
nrp.networkOutputNumb, // Support.GetOutputNumb(datasetPath);
denseActivations.Count,
denseActivations,
denseDropoutChB.Checked,
(int)denseDropoutNUD.Value);
var gp = new GeneticProgramm(nrp, crp, drp, (int)geneticEpochsNUD.Value, new[] { (int)copySelNUD.Value, (int)crossSelNUD.Value, (int)mutateRateNUD.Value },
(int)popolationCountNUD.Value, (int)mutateRateNUD.Value, (float)accPriorityNUD.Value, (float)memPriorityNUD.Value,
EstimatorCB.SelectedIndex, (int)instance.percentNUD.Value);
return(gp);
}