public void TestBackPropagationWithoutRegularisaion()
{
NeuralNetwork.CostFunctionWithThetaParameter backProp =
t =>
{
var backPropagationResult = NeuralNetwork.BackPropagation(Xm, ym, t, new List <int>()
{
4
}, 4, 0);
return(backPropagationResult);
};
var bothThetas = NeuralNetwork.PackThetas(new List <Matrix <double> >()
{
theta1NN, theta2NN
});
var resultNumericalGradient = NeuralNetwork.ComputeNumericalGradient(backProp, bothThetas);
var pathNumericalGradients = Path.Combine(Utils.GetAssemblyPath(), "..\\..\\TestData\\NumericalGradientForBackPropagationWithoutRegularisation.txt");
var matrixNumericalGradients = DelimitedReader.Read <double>(pathNumericalGradients);
Assert.IsTrue(MatricesEqual(resultNumericalGradient, matrixNumericalGradients));
var resultBackPropagation = NeuralNetwork.BackPropagation(Xm, ym, bothThetas, new List <int>()
{
4
}, 4, 0);
var pathGradientForBackPropagation = Path.Combine(Utils.GetAssemblyPath(), "..\\..\\TestData\\GradientForBackPropagationWithoutRegularisation.txt");
var matrixGradientForBackPropagation = DelimitedReader.Read <double>(pathGradientForBackPropagation);
Assert.IsTrue(Equalities.DoubleEquals(resultBackPropagation.Item1, 3.08744915815864));
Assert.IsTrue(MatricesEqual(resultBackPropagation.Item2, matrixGradientForBackPropagation));
Assert.IsTrue(MatricesEqual(resultBackPropagation.Item2, resultNumericalGradient, 0.000000001));
}
static void Main(string[] args)
{
Utils.SetNumberDecimalSeparatorToDotInCultureSettings();
var x_data = DelimitedReader.Read <double>(Path.Combine(Utils.GetAssemblyPath(), "..\\..\\TestData\\x_data.txt"));
var y_data = DelimitedReader.Read <double>(Path.Combine(Utils.GetAssemblyPath(), "..\\..\\TestData\\y_data.txt"));
NeuralNetwork.CostFunctionWithThetaParameter backProp =
tt =>
{
var backPropagationResult = NeuralNetwork.BackPropagation(x_data, y_data, tt, new List <int>()
{
25
}, 10, 0);
return(backPropagationResult);
};
var thetas = NeuralNetwork.RandomInitialiseWeights(400, 10, new List <int>()
{
25
});
var theta = NeuralNetwork.PackThetas(thetas);
var resultGradientDescent = NeuralNetwork.GradientDescent(backProp, theta, 1, 300);
DelimitedWriter.Write("..\\..\\TestData\\ThetaGradient.txt", resultGradientDescent.Item1, "thetaGradient");
DelimitedWriter.Write("..\\..\\TestData\\JHistory.txt", resultGradientDescent.Item2, "JHistory");
var resultTheta = resultGradientDescent.Item1;
//var resultComputeNumericalGradient = NeuralNetwork.ComputeNumericalGradient(backProp, resultTheta);
//resultComputeNumericalGradient.Save("D:/test/numericalGradients.txt");
var result5 = NeuralNetwork.BackPropagation(x_data, y_data, resultTheta, new List <int>(25), 10, 0);
var JJ = result5.Item1;
DelimitedWriter.Write("..\\..\\TestData\\GradientForBackPropagation.txt", result5.Item2, "gradBackPropagation");
var resultThetaLoadedFromFile = (DenseMatrix)DelimitedReader.Read <double>("..\\..\\TestData\\thetaGradient.txt");
var thetasList = NeuralNetwork.UnpackThetas(resultThetaLoadedFromFile, 400, new List <int>()
{
25
}, 10);
var result = NeuralNetwork.GetPredictions(x_data, y_data, thetasList);
Console.WriteLine(result.Item1);
DelimitedWriter.Write("..\\..\\TestData\\Predictions.txt", result.Item2, "");
}
public void TestBackPropagationWithRegularisation()
{
NeuralNetwork.CostFunctionWithThetaParameter backProp =
t =>
{
var backPropagationResult = NeuralNetwork.BackPropagation(Xm, ym, t, new List <int>()
{
4
}, 4, 1);
return(backPropagationResult);
};
var bothThetas = NeuralNetwork.PackThetas(new List <Matrix <double> >()
{
theta1NN, theta2NN
});
var resultNumericalGradient = NeuralNetwork.ComputeNumericalGradient(backProp, bothThetas);
var pathNumericalGradients = Path.Combine(Utils.GetAssemblyPath(), "..\\..\\TestData\\NumericalGradients.txt");
var matrixNumericalGradients = DelimitedReader.Read <double>(pathNumericalGradients);
Assert.IsTrue(MatricesEqual(resultNumericalGradient, matrixNumericalGradients));
var resultBackPropagation = NeuralNetwork.BackPropagation(Xm, ym, bothThetas, new List <int>()
{
4
}, 4, 1);
var pathGradientForBackPropagation = Path.Combine(Utils.GetAssemblyPath(), "..\\..\\TestData\\GradientForBackPropagation.txt");
var matrixGradientForBackPropagation = DelimitedReader.Read <double>(pathGradientForBackPropagation);
Assert.IsTrue(Equalities.DoubleEquals(resultBackPropagation.Item1, 3.46051055642594));
Assert.IsTrue(MatricesEqual(resultBackPropagation.Item2, matrixGradientForBackPropagation));
Assert.IsTrue(MatricesEqual(resultBackPropagation.Item2, resultNumericalGradient, 0.000000001));
var resultGradientDescent = NeuralNetwork.GradientDescent(backProp, bothThetas, 1, 3000);
var pathResultTheta = Path.Combine(Utils.GetAssemblyPath(), "..\\..\\TestData\\ThetaAfterGradientDescentForBackProp.txt");
var matrixResultTheta = DelimitedReader.Read <double>(pathResultTheta);
Assert.IsTrue(MatricesEqual(resultGradientDescent.Item1, matrixResultTheta));
var resultCost = resultGradientDescent.Item2.ToRowWiseArray().LastOrDefault(elem => elem != 0.0d);
Assert.IsTrue(Equalities.DoubleEquals(resultCost, 2.2493405784756875));
}
