private Vector3 GetRandomShift()
{
float x = RandomGaussian.Next(0.0f, deviation);
float z = RandomGaussian.Next(0.0f, deviation);
return(new Vector3(x, 0, z));
}
public static Network BuildNetwork(Random random,
ICostFunction costFunction, IRegularizationFunction regularizationFunction,
WeightIntializerType weightIntializerType,
DropoutLayerOptions dropoutLayerOptions,
int inputNeuronCount, int outputNeuronCount, params int[] hiddenLayerCounts)
{
Network network = new Network(costFunction, regularizationFunction,
dropoutLayerOptions, random);
network.InputLayer = InputLayer.BuildInputLayer(null, inputNeuronCount, random);
Layer previousLayer = network.InputLayer;
int dropoutLayerIndex = 1;
bool isDropoutLayer = false;
IWeightBuilder weightBuilder = null;
for (int c = 0; c < hiddenLayerCounts.Length; c++)
{
isDropoutLayer = dropoutLayerOptions.DropoutLayerIndices.Contains(dropoutLayerIndex);
int currentLayerCount = hiddenLayerCounts[c];
switch (weightIntializerType)
{
case WeightIntializerType.RandomGaussianWithNeuronCount:
weightBuilder = new RandomGaussianWithNeuronCount(previousLayer.Neurons.Count, 0, random);
break;
case WeightIntializerType.RandomNormal:
weightBuilder = new RandomGaussian(0, 1, random);
break;
}
HiddenLayer hiddenLayer = HiddenLayer.BuildHiddenLayer(weightBuilder, previousLayer,
currentLayerCount, isDropoutLayer ? dropoutLayerOptions.ProbabilityOfDropout : 0, random);
network.HiddenLayers.Add(hiddenLayer);
previousLayer = hiddenLayer;
dropoutLayerIndex++;
}
isDropoutLayer = dropoutLayerOptions.DropoutLayerIndices.Contains(dropoutLayerIndex);
switch (weightIntializerType)
{
case WeightIntializerType.RandomGaussianWithNeuronCount:
weightBuilder = new RandomGaussianWithNeuronCount(previousLayer.Neurons.Count, 0, random);
break;
case WeightIntializerType.RandomNormal:
weightBuilder = new RandomGaussian(0, 1, random);
break;
}
network.OutputLayer = OutputLayer.BuildOutputLayer(weightBuilder, (HiddenLayer)previousLayer,
outputNeuronCount, isDropoutLayer ? dropoutLayerOptions.ProbabilityOfDropout : 0, random);
return(network);
}
private Vector3 findSpreadDirection(Vector3 aim)
{
float spread1 = RandomGaussian.GetGaussian(0.0f, 2.0f, -horizontalSpread / 2.0f, horizontalSpread / 2.0f);
float spread2 = RandomGaussian.GetGaussian(0.0f, 2.0f, -verticalSpread / 2.0f, verticalSpread / 2.0f);
Vector3 retAim = Quaternion.Euler(spread1, spread2, spread1) * aim;
retAim.Normalize();
return(retAim);
}
private int[,] GenerateEnzymeArray(int numGroups, int numEnz)
{
int[,] protVals = new int[numGroups, numEnz];
// healthy baseline, normally distributed
int mean = 15; int sigma = 3;
RandomGaussian randGaussian = new RandomGaussian(sigma, mean);
// disease difference
float fractionDiff = .1f; // fraction of total enzyme types that are going to be different
int numDiff = Mathf.CeilToInt(Global.sim.limit_enzs * fractionDiff); // number of enzyme types that are different
int delta = 1;
for (int g = 0; g < numGroups; g++)
{
int[] ei = Global.RandPerm(numEnz);
for (int e = 0; e < numEnz; e++)
{
int value;
if (g == 0)
{
value = Mathf.Max(0, Mathf.RoundToInt(randGaussian.Get()));
} // healthy baseline
else if (e < numDiff)
{
value = protVals[0, ei[e]] + RandSign(delta);
} // perturbation
else
{
value = protVals[0, ei[e]];
} // no perturbation
protVals[g, ei[e]] = Mathf.Max(0, value);
}
}
return(protVals);
}
private float GetRawChangeLength(int min, int length)
{
// return (int)Math.Round(RandomGaussian.NextGaussian(toMinMax*_meanRange, (float)Math.Sqrt(toMinMax), 0f, toMinMax)); //doubtful about sqrt
return(RandomGaussian.NextGaussian(length * _meanRange, (length - min) * 0.2f, min, length));
}