public FullyConnectedLayer(int neurons, Layer previousLayer, ActivationFunctions.ActivationFunction activation)
: base(neurons, 1, 1, previousLayer)
{
Activation = activation;
this.weights = new float[Neurons * PreviousLayer.Neurons];
this.biases = new float[Neurons];
}
public ConvolutionalLayer(int filterCount, int filterSize, int stride, int zeroPadding, Layer previousLayer, ActivationFunctions.ActivationFunction activation)
: base(CalculateWidth(filterSize, stride, zeroPadding, previousLayer), CalculateHeight(filterSize, stride, zeroPadding, previousLayer), filterCount, previousLayer)
{
Activation = activation;
FilterSize = filterSize;
Stride = stride;
ZeroPadding = zeroPadding;
this.weights = new float[filterCount * filterSize * filterSize * previousLayer.Depth];
this.biases = new float[filterCount];
}
private static NeuralNetwork InitializeNeuralNetwork(int seed, ActivationFunctions.ActivationFunction activationFunction)
{
Random random = new Random(seed == 0 ? new Random().Next() : seed);
float RandomWeight() => (float)(random.NextDouble() * 2 - 1);
InputLayer l0 = new InputLayer(2);
FullyConnectedLayer l1 = new FullyConnectedLayer(2, l0, activationFunction);
for (int i = 0; i < l1.Weights; i++)
{
l1.SetWeight(i, RandomWeight() * 0.25f);
}
FullyConnectedLayer l2 = new FullyConnectedLayer(1, l1, activationFunction);
for (int i = 0; i < l2.Weights; i++)
{
l2.SetWeight(i, RandomWeight() * 0.25f);
}
return(new NeuralNetwork(l0, l1, l2));
}
public Dense(int units, ActivationFunctions.ActivationFunction actFunction)
{
activationFunction = actFunction;
outputShape = new int[] { units, 1 };
}
