public MultiLayerNeuralNet(MultiLayerNeuralNet pattern)
{
// Copy input layer wages.
this.wagesBetweenInputAndFirstHiddenLayer =
new float
[pattern.wagesBetweenInputAndFirstHiddenLayer.GetLength(0),
pattern.wagesBetweenInputAndFirstHiddenLayer.GetLength(1)];
Array.Copy(
pattern.wagesBetweenInputAndFirstHiddenLayer,
this.wagesBetweenInputAndFirstHiddenLayer,
pattern.wagesBetweenInputAndFirstHiddenLayer.GetLength(0) * pattern.wagesBetweenInputAndFirstHiddenLayer.GetLength(1));
//Copy hidden layers wages and biases.
this.wagesBetweenHiddenLayers = new float[pattern.wagesBetweenHiddenLayers.GetLength(0)][, ];
for (int i = 0; i < this.wagesBetweenHiddenLayers.Length; i++)
{
this.wagesBetweenHiddenLayers[i] =
new float[pattern.wagesBetweenHiddenLayers[i].GetLength(0), pattern.wagesBetweenHiddenLayers[i].GetLength(1)];
Array.Copy(
pattern.wagesBetweenHiddenLayers[i],
this.wagesBetweenHiddenLayers[i],
pattern.wagesBetweenHiddenLayers[i].GetLength(0) * pattern.wagesBetweenHiddenLayers[i].GetLength(1));
}
this.biasesInHiddenLayers = new float[pattern.biasesInHiddenLayers.Length][];
for (int i = 0; i < this.biasesInHiddenLayers.Length; i++)
{
this.biasesInHiddenLayers[i] = new float[pattern.biasesInHiddenLayers[i].Length];
Array.Copy(
pattern.biasesInHiddenLayers[i],
this.biasesInHiddenLayers[i],
pattern.biasesInHiddenLayers[i].Length);
}
// Copy output layer wages and biases.
this.wagesBetweenLastHiddenAndOutputLayer =
new float
[pattern.wagesBetweenLastHiddenAndOutputLayer.GetLength(0),
pattern.wagesBetweenLastHiddenAndOutputLayer.GetLength(1)];
Array.Copy(
pattern.wagesBetweenLastHiddenAndOutputLayer,
this.wagesBetweenLastHiddenAndOutputLayer,
pattern.wagesBetweenLastHiddenAndOutputLayer.GetLength(0) * pattern.wagesBetweenLastHiddenAndOutputLayer.GetLength(1));
this.biasesInOutputLayer = new float[pattern.biasesInOutputLayer.Length];
Array.Copy(
pattern.biasesInOutputLayer,
this.biasesInOutputLayer,
pattern.biasesInOutputLayer.Length);
// Copy activation function
this.activationFunction = pattern.activationFunction;
}
public INeuralNet Crossover(INeuralNet other)
{
if (other is MultiLayerNeuralNet == false)
{
throw new ApplicationException("'other' should be of type COneLayerNeuralNet to be able to crossover with COneLayerNeuralNet.");
}
var child = new MultiLayerNeuralNet(
this.wagesBetweenInputAndFirstHiddenLayer.GetLength(0),
this.wagesBetweenLastHiddenAndOutputLayer.GetLength(1),
this.wagesBetweenInputAndFirstHiddenLayer.GetLength(1),
this.wagesBetweenHiddenLayers.Length + 1);
var castedOther = other as MultiLayerNeuralNet;
child.wagesBetweenInputAndFirstHiddenLayer =
CrossoverHelper.Crossover(this.wagesBetweenInputAndFirstHiddenLayer, castedOther.wagesBetweenInputAndFirstHiddenLayer);
for (int i = 0; i < this.wagesBetweenHiddenLayers.Length; i++)
{
child.wagesBetweenHiddenLayers[i] =
CrossoverHelper.Crossover(this.wagesBetweenHiddenLayers[i], castedOther.wagesBetweenHiddenLayers[i]);
}
for (int i = 0; i < this.biasesInHiddenLayers.Length; i++)
{
child.biasesInHiddenLayers[i] =
CrossoverHelper.Crossover(this.biasesInHiddenLayers[i], castedOther.biasesInHiddenLayers[i]);
}
child.wagesBetweenLastHiddenAndOutputLayer =
CrossoverHelper.Crossover(this.wagesBetweenLastHiddenAndOutputLayer, castedOther.wagesBetweenLastHiddenAndOutputLayer);
child.biasesInOutputLayer =
CrossoverHelper.Crossover(this.biasesInOutputLayer, castedOther.biasesInOutputLayer);
return(child);
}
