private static void train(int ep)
{
// the input values
double[,] inputs =
{
{ 0, 0 },
{ 0, 1 },
{ 1, 0 },
{ 1, 1 }
};
// desired results
double[] results = { 0, 1, 1, 0 };
// creating the neurons
Neuron hiddenNeuron1 = new Neuron();
Neuron hiddenNeuron2 = new Neuron();
Neuron hiddenNeuron3 = new Neuron();
OutputNeuron outputNeuron = new OutputNeuron();
// random weights
hiddenNeuron1.randomizeWeights();
hiddenNeuron2.randomizeWeights();
hiddenNeuron3.randomizeWeights();
outputNeuron.randomizeWeights();
int epoch = 0;
Retry:
epoch++;
for (int i = 0; i < 4; i++) // very important, do NOT train for only one example
{
// 1) forward propagation (calculates output)
hiddenNeuron1.inputs = new double[] { inputs[i, 0], inputs[i, 1] };
hiddenNeuron2.inputs = new double[] { inputs[i, 0], inputs[i, 1] };
outputNeuron.inputs = new double[] { hiddenNeuron1.output, hiddenNeuron2.output };
Console.WriteLine("{0} xor {1} = {2}", inputs[i, 0], inputs[i, 1], outputNeuron.output());
Console.WriteLine(outputNeuron.output()[0]);
Console.WriteLine(outputNeuron.output()[1]);
Console.WriteLine(outputNeuron.output()[2]);
// 2) back propagation (adjusts weights)
// adjusts the weight of the output neuron, based on its error
outputNeuron.error = sigmoid.derivative(outputNeuron.output) * (results[i] - outputNeuron.output);
//get { return sigmoid.output(weights[0] * inputs[0] + weights[1] * inputs[1] + biasWeight); }
outputNeuron.adjustWeights();
// then adjusts the hidden neurons' weights, based on their errors
hiddenNeuron1.error = sigmoid.derivative(hiddenNeuron1.output) * outputNeuron.error * outputNeuron.weights[0];
hiddenNeuron2.error = sigmoid.derivative(hiddenNeuron2.output) * outputNeuron.error * outputNeuron.weights[1];
hiddenNeuron1.adjustWeights();
hiddenNeuron2.adjustWeights();
}
if (epoch < ep)
{
goto Retry;
}
OutPutNeuron1 = outputNeuron;
HiddenNeuron1 = hiddenNeuron1;
HiddenNeuron2 = hiddenNeuron2;
Console.ReadLine();
}
private static void train()
{
// the input values
double[] inputs = { 0.10, 0.06, 0.25, 0.16, 0.81, 1f, 0.36, 0.49, 0.01, 0.04, 1.21, 1.44, 1.69 };
// desired results
double[] results = { 0.023026, 0.0179176, 0.03218876, 0.0277259, 0.0439445, 0.0460517, 0.035835189, 0.03891820298, 0, 0.0138629, 0.04795791, 0.049698133, 0.05129899 };
// creating the neurons
Neuron N1 = new Neuron();
Neuron N2 = new Neuron();
OutputNeuron O1 = new OutputNeuron();
OutputNeuron O2 = new OutputNeuron();
OutputNeuron Output = new OutputNeuron();
// random weights
N1.randomizeWeights();
N2.randomizeWeights();
O1.randomizeWeights();
O2.randomizeWeights();
Output.randomizeWeights();
int epoch = 0;
Retry:
epoch++;
for (int i = 0; i < 13; i++) // very important, do NOT train for only one example
{
// 1) forward propagation (calculates output)
N1.input = inputs[i];
N2.input = inputs[i];
O1.inputs[0] = N1.output;
O1.inputs[1] = N2.output;
O2.inputs[0] = N1.output;
O2.inputs[1] = N2.output;
Output.inputs[0] = O1.output;
Output.inputs[1] = O2.output;
Console.WriteLine("Entrada: {0} Salida: {1}", 100 * N1.input, 100 * Output.output);
// 2) back propagation (adjusts weights)
// adjusts the weight of the output neuron, based on its error
Output.error = sigmoid.derivative(Output.output) * (results[i] - Output.output);
Output.adjustWeights();
// then adjusts the hidden neurons' weights, based on their errors
O1.error = sigmoid.derivative(O1.output) * O1.error * ((Output.weights[0] + Output.weights[1]) / 2);
O2.error = sigmoid.derivative(O2.output) * O2.error * ((Output.weights[0] + Output.weights[1]) / 2);
N1.error = sigmoid.derivative(N1.output) * N1.error * ((O1.weights[0] + O1.weights[1]) / 2);
N2.error = sigmoid.derivative(N2.output) * O2.error * ((O2.weights[0] + O2.weights[1]) / 2);
N1.adjustWeights();
N2.adjustWeights();
O1.adjustWeights();
O2.adjustWeights();
Output.adjustWeights();
}
if (epoch < 200000)
{
goto Retry;
}
Console.ReadLine();
bool salir = false;
while (salir == false)
{
Console.WriteLine("Introduzca una entrada:");
float entrada = Convert.ToSingle(Console.ReadLine());
N1.input = entrada / 100;
N2.input = entrada / 100;
O1.inputs[0] = N1.output;
O1.inputs[1] = N2.output;
O2.inputs[0] = N1.output;
O2.inputs[1] = N2.output;
Output.inputs[0] = O1.output;
Output.inputs[1] = O2.output;
Console.WriteLine("Salida: " + 100 * Output.output);
}
}
