public void MeanSquareError_ThrowOnLengthMismatch()
{
var target = new[] { 1.0, 1.0 };
var output = new[] { 1.0 };
Assert.Throws <NeuralNetworkException>(() => ErrorCalculations.MeanSquareError(target, output));
}
public void CrossEntropyError_Calculate()
{
var target = new[] { 0.05, 0.8, 0.15 };
var output = new[] { 0.3, 0.5, 0.2 };
var actual = ErrorCalculations.CrossEntropyError(target, output);
var expected = 0.856132071529368;
Assert.Equal(expected, actual, 12);
}
public void MeanSquareError_Calculate()
{
var target = new[] { 0.05, 0.8, 0.15 };
var output = new[] { 0.3, 0.5, 0.2 };
var actual = ErrorCalculations.MeanSquareError(target, output);
var expected = 0.0516666666666667;
Assert.Equal(expected, actual, 12);
}
public ErrorsViewModel(Types.Signal original, Types.Signal processed, int neighboursCount, double samplingFrequency)
{
_neighboursCount = neighboursCount;
_samplingFrequency = samplingFrequency;
_meanSquaredError = ErrorCalculations.meanSquaredError(original.points, processed.points);
_signalToNoiseRatio = ErrorCalculations.signalToNoiseRatio(original.points, processed.points);
_peakSignalToNoiseRatio = ErrorCalculations.peakSignalToNoiseRatio(original.points, processed.points);
_maximumDifference = ErrorCalculations.maxDifference(original.points, processed.points);
SaveResultsCommand = new RelayCommand(SaveResults);
RememberResultsCommand = new RelayCommand(RememberResults);
}
public static double GetError(INeuralNetwork nn, IList <InputOutput> testSet)
{
var error = 0.0;
foreach (var inputOutput in testSet)
{
var result = nn.FeedForward(inputOutput.Input);
error += ErrorCalculations.CrossEntropyError(inputOutput.Output, result.Output);
}
return(error / testSet.Count);
}
public static double GetMeanSquareError(double[] expected, double[] actual)
{
return(ErrorCalculations.MeanSquareError(expected, actual));
}
public static double GetCrossEntropyError(double[] expected, double[] actual)
{
return(ErrorCalculations.CrossEntropyError(expected, actual));
}
