Functions
def	RandomSamplesUniform (minCoordinate, maxCoordinate, dim, numSamples)

def	RandomSamplesGauss (mean, sigma, dim, numSamples)

def	ExactFunction (dimOutput, coordinates)

Variables
int	randomSeed = 1234567

int	minCoordinate = -3

int	maxCoordinate = 3

int	meanNoise = 0

int	sigmaNoise = 1

int	dimInput = 2

int	dimOutput = 2

int	numSamples = 10

int	numNeuronsHiddenLayer = 3

int	verboseLevel = 0

	SupportPointsInput = RandomSamplesUniform(minCoordinate, maxCoordinate, dimInput, numSamples)

	SupportPointsOutputExact = ExactFunction(dimOutput, SupportPointsInput)

	SupportPointsOutputNoise = RandomSamplesGauss(meanNoise, sigmaNoise, dimOutput, numSamples)

	SupportPointsOutputWithNoise = SupportPointsOutputExact+SupportPointsOutputNoise

	myNetwork = nuto.NeuralNetwork([numNeuronsHiddenLayer])

int	lowerBound = -1

int	upperBound = 1

	SupportPointsApproximation = np.zeros((dimOutput, numSamples))

	SupportPointsApproximationMin = np.zeros((dimOutput, numSamples))

	SupportPointsApproximationMax = np.zeros((dimOutput, numSamples))

	NumParameters = myNetwork.GetNumParameters()

	gradient = np.zeros((NumParameters, 1))

	hessian = np.zeros((NumParameters, NumParameters))

Function Documentation

def NeuralNetwork.ExactFunction	(	dimOutput,
		coordinates
	)

def NeuralNetwork.RandomSamplesGauss	(	mean,
		sigma,
		dim,
		numSamples
	)

def NeuralNetwork.RandomSamplesUniform	(	minCoordinate,
		maxCoordinate,
		dim,
		numSamples
	)

Variable Documentation

int NeuralNetwork.dimInput = 2

int NeuralNetwork.dimOutput = 2

NeuralNetwork.gradient = np.zeros((NumParameters, 1))

NeuralNetwork.hessian = np.zeros((NumParameters, NumParameters))

int NeuralNetwork.lowerBound = -1

int NeuralNetwork.maxCoordinate = 3

int NeuralNetwork.meanNoise = 0

int NeuralNetwork.minCoordinate = -3

NeuralNetwork.myNetwork = nuto.NeuralNetwork([numNeuronsHiddenLayer])

int NeuralNetwork.numNeuronsHiddenLayer = 3

NeuralNetwork.NumParameters = myNetwork.GetNumParameters()

int NeuralNetwork.numSamples = 10

int NeuralNetwork.randomSeed = 1234567

int NeuralNetwork.sigmaNoise = 1

NeuralNetwork.SupportPointsApproximation = np.zeros((dimOutput, numSamples))

NeuralNetwork.SupportPointsApproximationMax = np.zeros((dimOutput, numSamples))

NeuralNetwork.SupportPointsApproximationMin = np.zeros((dimOutput, numSamples))

NeuralNetwork.SupportPointsOutputExact = ExactFunction(dimOutput, SupportPointsInput)

int NeuralNetwork.upperBound = 1

int NeuralNetwork.verboseLevel = 0