private void runIrisNetwork()
{
//Getting data set ready:
bool normalize = selectedNetwork == typeof(LMS);
irisSet = new DataSetReader("../../../DataSets/iris.data", DataSetType.IRIS, normalize);
graphDrawer = new GraphDrawer(irisSet, graphPictureBox, irisBrushes);
//Network parameters:
target = new double[] { -1, 1 };
features = new double[] { (int)numFeatureOne.Value, (int)numFeatureTwo.Value };
classes = new double[] { (int)numClassOne.Value, (int)numClassTwo.Value };
double eta = (double)numEta.Value;
double bias = 1.0;
//Selecting the right network:
if (selectedNetwork == typeof(Perceptron))
{
machine = new Perceptron(irisSet, target, eta, bias, features, classes);
}
else if (selectedNetwork == typeof(LMS))
{
machine = new LMS(irisSet, target, eta, bias, features, classes);
}
//Train the network:
machine.train(30);
//Test the network:
int[,] testMatrix = machine.test(20);
UiTools.drawMatrix(dataGridView, testMatrix);
lblAccuracy.Text = VectorTools.confusionAccuracy(testMatrix).ToString();
}
public NeuralNetwork(DataSetReader dataSet, double[] target, double[] weight, double eta, double bias)
{
this.target = target;
this.data = dataSet.data;
this.featureMask = VectorTools.sequence(dataSet.features);
this.classMask = VectorTools.sequence(dataSet.classes);
this.samples = dataSet.samples;
this.weight = weight;
this.eta = eta;
this.bias = bias;
}
protected const int MAX_EPOCHS = 1000; //Limiting the number of iterations in the process.
/*
* CONSTRUCTORS
*/
public NeuralNetwork(DataSetReader dataSet, double[] target, double bias, double eta)
{
this.target = target;
this.data = dataSet.data;
this.featureMask = VectorTools.sequence(dataSet.features);
this.classMask = VectorTools.sequence(dataSet.classes);
this.weight = VectorTools.ones(this.featureMask.Length + 1); //+1 for bias.
this.samples = dataSet.samples;
this.eta = eta;
this.bias = bias;
}
public NeuralNetwork(DataSetReader dataSet, double[] target, double eta, double bias, double[] featureMask, double[] classMask)
{
if (featureMask.Length > dataSet.features || featureMask.Length < 1)
{
throw new ArgumentOutOfRangeException("Number of features specified is not applicable");
}
this.target = target;
this.data = dataSet.data;
this.featureMask = featureMask;
this.classMask = classMask;
this.weight = VectorTools.ones(this.featureMask.Length + 1); //+1 for bias.
this.samples = dataSet.samples;
this.eta = eta;
this.bias = bias;
}
public LMS(DataSetReader dataSet, double[] target, double[] weight, double eta, double bias)
: base(dataSet, target, weight, eta, bias)
{
checkNormalized(dataSet.normalized);
this.data = dataSet.dataNorm;
}
public LMS(DataSetReader dataSet, double[] target, double eta, double bias, double[] featureMask, double[] classMask)
: base(dataSet, target, eta, bias, featureMask, classMask)
{
checkNormalized(dataSet.normalized);
this.data = dataSet.dataNorm;
}
public GraphDrawer(DataSetReader dataSet, PictureBox pictureBox, SolidBrush[] brush)
{
this.dataSet = dataSet;
this.pictureBox = pictureBox;
this.brush = brush;
}
/// <summary>Create a new perceptron machine.</summary>
/// <param name="dataSet">Object holding the dataset to work on.</param>
/// <param name="target">The target/goal vector for the training.</param>
/// <param name="weight">The initial weight vector to start with.</param>
/// <param name="eta">The learning rate.</param>
/// <param name="bias">Initial bias value.</param>
public Perceptron(DataSetReader dataSet, double[] target, double[] weight, double eta, double bias)
: base(dataSet, target, weight, eta, bias)
{
//Nothing here.
}
/// <summary>Create a new perceptron machine.</summary>
/// <param name="dataSet">Object holding the dataset to work on.</param>
/// <param name="target">The target/goal vector for the training.</param>
/// <param name="eta">The learning rate.</param>
/// <param name="bias">Initial bias value.</param>
/// <param name="featureMask">Set of features to use in the machine.</param>
public Perceptron(DataSetReader dataSet, double[] target, double eta, double bias, double[] featureMask, double[] classMask)
: base(dataSet, target, eta, bias, featureMask, classMask)
{
//Nothing here.
}
