/// <summary>
/// The SSE error with the current weights.
/// </summary>
/// <returns></returns>
private double CalculateError()
{
var result = new ErrorCalculation();
for (int i = 0; i < _trainingLength; i++)
{
_pair = _indexableTraining[i];
IMLData actual = _network.Compute(_pair.Input);
result.UpdateError(actual, _pair.Ideal, _pair.Significance);
}
return result.CalculateSSE();
}
/// <summary>
/// Called internally to compute each output neuron.
/// </summary>
/// <param name="outputNeuron">The output neuron to compute.</param>
private void InternalCompute(int outputNeuron)
{
int row = 0;
var error = new ErrorCalculation();
EngineArray.Fill(derivative, 0);
// Loop over every training element
foreach (var pair in training)
{
var networkOutput = network.Compute(pair.Input);
double e = pair.Ideal.Data[outputNeuron] - networkOutput[outputNeuron];
error.UpdateError(networkOutput[outputNeuron], pair.Ideal[outputNeuron]);
int currentWeight = 0;
// loop over the output weights
int outputFeedCount = network.GetLayerTotalNeuronCount(network.LayerCount - 2);
for (int i = 0; i < network.OutputCount; i++)
{
for (int j = 0; j < outputFeedCount; j++)
{
double jc;
if (i == outputNeuron)
{
jc = ComputeDerivative(pair.Input, outputNeuron,
currentWeight, _dStep,
networkOutput[outputNeuron], row);
}
else
{
jc = 0;
}
gradients[currentWeight] += jc*e;
derivative[currentWeight] += jc;
currentWeight++;
}
}
// Loop over every weight in the neural network
while (currentWeight < network.Flat.Weights.Length)
{
double jc = ComputeDerivative(
pair.Input, outputNeuron, currentWeight,
_dStep,
networkOutput[outputNeuron], row);
derivative[currentWeight] += jc;
gradients[currentWeight] += jc*e;
currentWeight++;
}
row++;
}
UpdateHessian(derivative);
sse += error.CalculateSSE();
}