| public RunEpoch ( double input, double output ) : double | ||
| input | double | Array of input vectors. |
| output | double | Array of output vectors. |
| return | double | |
public void test() {
// initialize input and output values
var input = new double[4][] {
new double[] { 0, 0 }, new double[] { 0, 1 },
new double[] { 1, 0 }, new double[] { 1, 1 }
};
var output = new double[4][] {
new double[] { 0 }, new double[] { 1 },
new double[] { 1 }, new double[] { 1 }
};
// create neural network
var network = new ActivationNetwork(
new SigmoidFunction(2),
2, // two inputs in the network
//2, // two neurons in the first layer
1); // one neuron in the second layer
// create teacher
var teacher =
new BackPropagationLearning(network);
// loop
while (true) {
// run epoch of learning procedure
var error = teacher.RunEpoch(input, output);
// check error value to see if we need to stop
// ...
if (error < 0.001) {
break;
}
}
Console.WriteLine(network.Compute(new double[] { 0, 0 })[0] + ","
+ network.Compute(new double[] { 0, 1 })[0] + ","
+ network.Compute(new double[] { 1, 0 })[0] + ","
+ network.Compute(new double[] { 1, 1 })[0]);
}
public void Test()
{
ActivationNetwork network = new ActivationNetwork(
new SigmoidFunction(),
2, // two inputs in the network
2, // two neurons in the first layer
1); // one neuron in the second layer
BackPropagationLearning teacher = new BackPropagationLearning(network);
double lastError = double.MaxValue;
int counter = 0;
while (true)
{
counter++;
var error = teacher.RunEpoch(input, output);
if (lastError - error < 0.0000001 && error < 0.001)
break;
lastError = error;
}
//var bla = network.Compute(input[0])[0];
//var round = Math.Round(network.Compute(input[0])[0], 2);
//var result = output[0][0];
//Assert.IsTrue(Math.Abs(round - result) < double.Epsilon);
Assert.IsTrue(Math.Abs(network.Compute(input[0])[0] - output[0][0]) < 0.03);
Assert.IsTrue(Math.Abs(network.Compute(input[1])[0] - output[1][0]) < 0.03);
Assert.IsTrue(Math.Abs(network.Compute(input[2])[0] - output[2][0]) < 0.03);
Assert.IsTrue(Math.Abs(network.Compute(input[3])[0] - output[3][0]) < 0.03);
Console.WriteLine($"Loop counter = {counter}.");
}
static void Main(string[] args)
{
// initialize input and output values
double[][] input = new double[4][] {
new double[] {0, 0}, new double[] {0, 1},
new double[] {1, 0}, new double[] {1, 1}
};
double[][] output = new double[4][] {
new double[] {0}, new double[] {1},
new double[] {1}, new double[] {0}
};
// create neural network
ActivationNetwork network = new ActivationNetwork(
new SigmoidFunction(1),
2, // two inputs in the network
2, // two neurons in the first layer
1); // one neuron in the second layer
// create teacher
BackPropagationLearning teacher =
new BackPropagationLearning(network);
// loop
for (int i = 0; i < 10000; i++)
{
// run epoch of learning procedure
double error = teacher.RunEpoch(input, output);
// check error value to see if we need to stop
// ...
Console.Out.WriteLine("#" + i + "\t" + error);
}
double[] ret1 = network.Compute(new double[] { 0, 0 });
double[] ret2 = network.Compute(new double[] { 1, 0 });
double[] ret3 = network.Compute(new double[] { 0, 1 });
double[] ret4 = network.Compute(new double[] { 1, 1 });
Console.Out.WriteLine();
Console.Out.WriteLine("Eval(0, 0) = " + ret1[0]);
Console.Out.WriteLine("Eval(1, 0) = " + ret2[0]);
Console.Out.WriteLine("Eval(0, 1) = " + ret3[0]);
Console.Out.WriteLine("Eval(1, 1) = " + ret4[0]);
Console.ReadLine();
}
public string Learn(string knowFile)
{
var network = new ActivationNetwork(new BipolarSigmoidFunction(), Constants.StoneCount, 1);
var teacher = new BackPropagationLearning(network); //new PerceptronLearning(network);
var data = LoadData(knowFile);
double error = int.MaxValue;
int index = 0;
while (error > 1.0 && index++ < 5000) {
error = teacher.RunEpoch(data.Item1, data.Item2);
}
var networkFile = knowFile + ".net";
network.Save(networkFile);
Console.WriteLine("Learn: {0}, Gen: {1}", knowFile, networkFile);
return networkFile;
}
public NeuralNetworkOperations(int characterSize)
{
neuralNet = new ActivationNetwork(new BipolarSigmoidFunction(2.0f), characterSize, 400, characterCount);
neuralNet.Randomize();
teacher = new AForge.Neuro.Learning.BackPropagationLearning(neuralNet);
teacher.LearningRate = 0.5f;
teacher.Momentum = 0.1f;
prepareDataForTeacher();
//var letters = treningLetterListInput.Zip(treningLetterListOutput, (i,o) => new { treningLetterListInput = i, treningLetterListOutput = o });
double err = 400.0f;
int count = 0;
while(err >= 30.0f)
{
err = teacher.RunEpoch(treningLetterListInput.ToArray(), treningLetterListOutput.ToArray());
count++;
}
}
public void Test()
{
var network = new ActivationNetwork(new SigmoidFunction(), inputCount, firstLayerNeurons, secondLayerNeurons, thirdLayerNeurons, lastLayerNeurons);
var teacher = new BackPropagationLearning(network);
var lastError = double.MaxValue;
var counter = 0;
while (true)
{
counter++;
var error = teacher.RunEpoch(input, output);
if ((lastError - error < 0.00001 && error < 0.01) || counter > 1200000)
break;
lastError = error;
}
var result1 = network.Compute(new double[] {1, 0, 1, 0, 1, 0, 1, 0});
Console.WriteLine($"2 + 2, 2 * 2 = {result1[0]}, {result1[1]}");
var result2 = network.Compute(new double[] {0, 1, 0, 1, 1, 0, 0, 1});
Console.WriteLine($"1 + 1, 2 * 1 = {result2[0]}, {result2[1]}");
var result3 = network.Compute(new double[] {1, 0, 1, 0, 0, 1, 0, 0});
Console.WriteLine($"2 + 2, 1 * 0 = {result3[0]}, {result3[1]}");
var result4 = network.Compute(new double[] {0, 1, 0, 0, 0, 1, 1, 0});
Console.WriteLine($"1 + 0, 1 * 2 = {result4[0]}, {result4[1]}");
}
/*
* Naucz sieć neuronową.
* Uczy perceptron przy pomocy algorytmu wstecznej propagacji.
* Sieć oczy się doputy, dopuki błąd uczenia w danej epoce nie będzie mniejszy niż
* dany margines błędu (this.errorRate). Zwraca ilość epok nauczania.
* double [][] input - tablica wektorów wejściowych
* double [][] output - tablica oczekiwanych wektorów wyjściowych
*/
public int Learn(double [][] input, double [][] output)
{
if (input.Length == output.Length)
{
double samples = (double)input.Length;
this.network = new ActivationNetwork(new BipolarSigmoidFunction(this.sigmoidAlphaValue), input[0].Length, this.neuronsInFirstLayer, 1);
BackPropagationLearning teacher = new BackPropagationLearning(network);
teacher.LearningRate = this.learningRate;
teacher.Momentum = this.momentum;
int epoch = 0;
while (teacher.RunEpoch(input, output) < this.errorRate) { epoch++; }
return epoch;
}
else
{
throw new Exception("Size of input and output arrays differs!");
}
}
static void Learn()
{
var network = new ActivationNetwork(
new SigmoidFunction(),
baseMaker.InputSize,
arguments.NeuronsCount,
baseMaker.OutputSize
);
network.Randomize();
foreach (var l in network.Layers)
foreach (var n in l.Neurons)
for (int i = 0; i < n.Weights.Length; i++)
n.Weights[i] = rnd.NextDouble() * 2 - 1;
var teacher = new BackPropagationLearning(network);
teacher.LearningRate = 1;
teacher.Momentum = 0;
while (true)
{
var watch = new Stopwatch();
watch.Start();
while (watch.ElapsedMilliseconds < 500)
{
teacher.RunEpoch(baseMaker.Inputs, baseMaker.Answers);
}
watch.Stop();
var count = 0;
percentage = new double[baseMaker.OutputSize, baseMaker.OutputSize];
for (int i = 0; i < baseMaker.OutputSize; i++)
for (int j = 0; j < baseMaker.OutputSize * 5; j++)
{
var task = baseMaker.GenerateRandom(i);
var output = network.Compute(task);
var max = output.Max();
var maxIndex = Enumerable.Range(0, output.Length).Where(z => output[z] == max).First();
percentage[i, maxIndex]++;
if (i != maxIndex) totalErrors++;
count++;
}
var maxPercentage = percentage.Cast<double>().Max();
for (int i = 0; i < baseMaker.OutputSize; i++)
for (int j = 0; j < baseMaker.OutputSize; j++)
percentage[i, j] /= maxPercentage;
totalErrors /= count;
form.BeginInvoke(new Action(Update));
}
}
public void Train(List<Session> train, List<Session> cv, out List<double> trainErrors, out List<double> cvErrors, IActivationFunction function)
{
trainErrors = new List<double>();
cvErrors = new List<double>();
var count = train.Count;
// prepare learning data
Console.WriteLine("prepare learning data");
double[][] input = new double[count][];
double[][] output = new double[count][];
// preparing the data
for (int i = 0; i < count; i++)
{
input[i] = CreateInput(train[i]);
output[i] = CreateOutput(train[i]);
}
Console.WriteLine("feature scaling");
mean = new double[inputSize];
dev = new double[inputSize];
for (int i = 0; i < inputSize; i++)
{
var query = input.Select(p => p[i]);
mean[i] = query.Average();
dev[i] = query.Deviation(mean[i]);
}
for (int i = 0; i < count; i++)
for (int j = 0; j < inputSize; j++)
{
input[i][j] = (input[i][j] - mean[j]) / dev[j];
}
Console.WriteLine("prepare cv data");
// prepare cv data
double[][] cvIn = new double[cv.Count][];
double[][] cvOut = new double[cv.Count][];
// preparing the data
for (int i = 0; i < cv.Count; i++)
{
cvIn[i] = CreateInput(cv[i]);
cvOut[i] = CreateOutput(cv[i]);
}
Console.WriteLine("cv feature scaling");
for (int i = 0; i < cv.Count; i++)
cvIn[i] = ScaleInput(cvIn[i]);
Console.WriteLine("create network");
// create perceptron
_network = new ActivationNetwork(function, inputSize, inputSize, classesCount);
_network.Randomize();
// create teacher
//PerceptronLearning teacher = new PerceptronLearning(_network);
BackPropagationLearning teacher = new BackPropagationLearning(_network);
// set learning rate
teacher.LearningRate = 0.01;
// loop
int iter = 0;
double error = 999;
double delta = 999;
Console.WriteLine("Train Network");
//while (iter < 1000)
while (delta > 1 && iter < 5000)
//while (iter < 2000)
{
// run epoch of learning procedure
double trainError = teacher.RunEpoch(input, output);
double trainError2 = ComputeCVError(_network, input, output);
double cvError = ComputeCVError(_network, cvIn, cvOut);
delta = Math.Abs(error - trainError);
error = trainError;
trainErrors.Add(trainError2);
cvErrors.Add(cvError);
iter++;
if (iter % 100 == 0)
Console.WriteLine(iter);
}
Console.WriteLine(iter);
}
private void backgroundWorker_DoWork(object sender, DoWorkEventArgs e)
{
if (!(e.Argument is TrainingOptions))
{
backgroundWorker.ReportProgress(0, "Bad thread argument!");
e.Cancel = true;
return;
}
TrainingOptions options = (TrainingOptions)e.Argument;
//Create Teacher
BackPropagationLearning networkTeacher = new BackPropagationLearning(m_networkContainer.ActivationNetwork);
networkTeacher.LearningRate = options.firstLearningRate;
networkTeacher.Momentum = options.momentum;
//Start Training
bool stop = false;
int lastStatusEpoch = 0;
int lastGraphEpoch = 0;
int lastSaveEpoch = 0;
backgroundWorker.ReportProgress(0, "Training...");
while (!stop)
{
#region Training Epoch
this.m_networkState.ErrorTraining = networkTeacher.RunEpoch(options.TrainingVectors.Input, options.TrainingVectors.Output) /* / options.TrainingVectors.Input.Length */;
this.m_networkState.ErrorValidation = networkTeacher.MeasureEpochError(options.ValidationVectors.Input, options.ValidationVectors.Output) /* / options.ValidationVectors.Input.Length */;
// Adjust Training Rate
if (options.secondLearningRate.HasValue)
networkTeacher.LearningRate = options.secondLearningRate.Value;
#endregion
#region Mark Network Savepoint
if (Properties.Settings.Default.training_Autosave == true &&
m_networkState.Epoch >= lastSaveEpoch + Properties.Settings.Default.training_AutosaveEpochs)
{
backgroundWorker.ReportProgress(0,UpdateType.NetworkSave);
lastSaveEpoch = m_networkState.Epoch;
}
#endregion
#region Graph Update
if (Properties.Settings.Default.graph_Disable == false &&
m_networkState.Epoch >= lastGraphEpoch + Properties.Settings.Default.graph_UpdateRate)
{
backgroundWorker.ReportProgress(0,UpdateType.Graph);
lastGraphEpoch = m_networkState.Epoch;
}
#endregion
#region Statusbar Update
if (Properties.Settings.Default.display_UpdateByTime == false &&
m_networkState.Epoch >= lastStatusEpoch + Properties.Settings.Default.display_UpdateRate)
{
if (options.TrainingType == TrainingType.ByError)
{
if (m_networkState.ErrorTraining != 0)
m_networkState.Progress = Math.Max(Math.Min((int)((options.limError * 100) / m_networkState.ErrorTraining), 100), 0);
}
else
{
if (m_networkState.Epoch != 0)
m_networkState.Progress = Math.Max(Math.Min((int)((m_networkState.Epoch * 100) / options.limEpoch), 100), 0);
}
backgroundWorker.ReportProgress(0, UpdateType.Statusbar);
lastStatusEpoch = m_networkState.Epoch;
}
#endregion
++m_networkState.Epoch;
// Sleep thread according to specified delay
System.Threading.Thread.Sleep(Properties.Settings.Default.training_delay);
#region Stop Conditions
if (options.TrainingType == TrainingType.ByError)
{
if (m_networkState.ErrorTraining <= options.limError)
stop = true;
}
else if (options.TrainingType == TrainingType.ByEpoch)
{
if (m_networkState.Epoch >= options.limEpoch)
stop = true;
}
if (backgroundWorker.CancellationPending)
{
e.Cancel = true;
stop = true;
}
#endregion
}
backgroundWorker.ReportProgress(0);
}
// Worker thread
void SearchSolution()
{
// number of learning samples
int samples = data.GetLength(0);
// data transformation factor
double yFactor = 1.7 / chart.RangeY.Length;
double yMin = chart.RangeY.Min;
double xFactor = 2.0 / chart.RangeX.Length;
double xMin = chart.RangeX.Min;
// prepare learning data
double[][] input = new double[samples][];
double[][] output = new double[samples][];
for (int i = 0; i < samples; i++)
{
input[i] = new double[1];
output[i] = new double[1];
// set input
input[i][0] = (data[i, 0] - xMin) * xFactor - 1.0;
// set output
output[i][0] = (data[i, 1] - yMin) * yFactor - 0.85;
}
// create multi-layer neural network
ActivationNetwork network = new ActivationNetwork(
new BipolarSigmoidFunction(sigmoidAlphaValue),
1, neuronsInFirstLayer, 1);
// create teacher
BackPropagationLearning teacher = new BackPropagationLearning(network);
// set learning rate and momentum
teacher.LearningRate = learningRate;
teacher.Momentum = momentum;
// iterations
int iteration = 1;
// solution array
double[,] solution = new double[50, 2];
double[] networkInput = new double[1];
// calculate X values to be used with solution function
for (int j = 0; j < 50; j++)
{
solution[j, 0] = chart.RangeX.Min + (double)j * chart.RangeX.Length / 49;
}
// loop
while (!needToStop)
{
// run epoch of learning procedure
double error = teacher.RunEpoch(input, output) / samples;
// calculate solution
for (int j = 0; j < 50; j++)
{
networkInput[0] = (solution[j, 0] - xMin) * xFactor - 1.0;
solution[j, 1] = (network.Compute(networkInput)[0] + 0.85) / yFactor + yMin;
}
chart.UpdateDataSeries("solution", solution);
// calculate error
double learningError = 0.0;
for (int j = 0, k = data.GetLength(0); j < k; j++)
{
networkInput[0] = input[j][0];
learningError += Math.Abs(data[j, 1] - ((network.Compute(networkInput)[0] + 0.85) / yFactor + yMin));
}
// set current iteration's info
UpdateTextbox(currentIterationBox, iteration.ToString());
//currentIterationBox.Text = iteration.ToString();
UpdateTextbox(currentErrorBox, learningError.ToString("F3"));
//currentErrorBox.Text = learningError.ToString("F3");
// increase current iteration
iteration++;
// check if we need to stop
if ((iterations != 0) && (iteration > iterations))
break;
}
// enable settings controls
EnableControls(true);
}
private void Worker()
{
ExtendedIOHelpers.ShowAlert("Generate new samples before training? y/n ", () => GenerateSamples());
if (!LoadSamples())
{
return;
}
Console.WriteLine("Loaded {0} samples", _inputList.Count);
var network = new ActivationNetwork(new SigmoidFunction(1), InputCount, OutputCount);
var teacher = new BackPropagationLearning(network)
{
LearningRate = 0.01
};
Console.WriteLine("Start training the network.");
int iteration = 0;
const int iterations = 5000;
double error = 1.0;
var st = new Stopwatch();
st.Start();
while (iteration < iterations && error > 0.00005)
{
error = teacher.RunEpoch(_inputList.ToArray(), _outputList.ToArray()) / _inputList.Count;
iteration++;
}
var time = st.ElapsedMilliseconds / 1000.0;
st.Stop();
Console.WriteLine("Network successfully trained! Error = {0:0.######}, Iteration = {1}", error, iteration);
Console.WriteLine("Time = {0:0.000} s\n", time);
// Normalize weights and convert to string format.
var weights = network.Layers
.Select(layer => layer.Neurons
.Select(neuron => neuron.Weights
.Select(x => string.Format("{0,6}", Convert.ToInt32(x * 1000.0)))));
ExtendedIOHelpers.ShowAlert("Do you want to save network to file? y/n ", () =>
{
SaveWeights(weights);
network.Save(NetworkFile);
});
Console.ReadLine();
}
// teach the network
public void learn()
{
network = new ActivationNetwork(new SigmoidFunction(sigmoidAV), 4, 5, 4);
teacher = new BackPropagationLearning(network);
teacher.LearningRate = learningRate;
teacher.Momentum = momentum;
error = 0;
Random randomG = new Random();
for (int i = 0; i < iteration; i++)
{
if (randomMomentum)
{
double random = (randomG.NextDouble() * 0.3) + 0.5;
teacher.Momentum = random;
}
error = teacher.RunEpoch(input, output) / sampleCount;
if(outputErrorComponent != null)
{
outputErrorComponent.Text = "Neural network error is : " + error;
}
}
Console.WriteLine("error is " + error);
}
private void SearchSolution()
{
MemoryStream ms = new MemoryStream();
mNetwork = new ActivationNetwork(
new BipolarSigmoidFunction(sigmoidAlphaValue),
mInput[0].Length, mHiddenNeurons, mOutput[0].Length);
// create teacher
BackPropagationLearning teacher = new BackPropagationLearning(mNetwork);
// set learning rate and momentum
teacher.LearningRate = mLearningRate;
teacher.Momentum = mMomentum;
bool needToStop = false;
int iteration = 0;
while (!needToStop)
{
double error = teacher.RunEpoch(mInput, mOutput) / mInput.Length;
mErrors[iteration] = error;
double test_error = 0.0;
for (int i = 0; i < mTestInput.Length; i++)
{
double[] test_result = mNetwork.Compute(mTestInput[i]);
test_error += ( mTestOutput[i][0] - test_result[0])*( mTestOutput[i][0] - test_result[0]);
}
mTestErrors[iteration] = Math.Sqrt(test_error);
if (min_test_error > mTestErrors[iteration])
{
min_test_error = mTestErrors[iteration];
// mTestBestNetwork = mNetwork;
ms = new MemoryStream();
mNetwork.Save(this.Id + ".txt");
}
iteration++;
if (iteration >= mIterations) needToStop = true;
}
mTestBestNetwork = (ActivationNetwork)ActivationNetwork.Load(this.Id + ".txt");
}
public static void ExactTrainingData()
{
TrainData[] tdatas;
// AForge.Neuro.Learning.BackPropagationLearning beural=new AForge.Neuro.Learning.BackPropagationLearning(new AForge.Neuro.ActivationNetwork(
tdatas = traindatas.ToArray();
double[][] output = new double[tdatas.Length-1][];
double[][] input = new double[tdatas.Length-1][];
for (int i = 1; i < tdatas.Length; i++)
{
int spddiff = 0, voldiff = 0, occdiff = 0, u_spddifft = 0, u_voldifft = 0, u_occdifft = 0, d_spddifft = 0, d_voldifft = 0, d_occdifft = 0,level=0;
tdatas[i].getTrainData(ref voldiff, ref spddiff, ref occdiff);
u_spddifft = tdatas[i].vd1.AvgSpd - tdatas[i - 1].vd1.AvgSpd;
u_voldifft = tdatas[i].vd1.Volume - tdatas[i-1].vd1.Volume;
u_occdifft = tdatas[i].vd1.Occupancy - tdatas[i - 1].vd1.Occupancy;
d_spddifft = tdatas[i].vd2.AvgSpd - tdatas[i - 1].vd2.AvgSpd;
d_voldifft = tdatas[i].vd2.Volume - tdatas[i - 1].vd2.Volume;
d_occdifft = tdatas[i].vd2.Occupancy - tdatas[i - 1].vd2.Occupancy;
level = tdatas[i-1].Level;
output[i-1] = new double[1];
output[i-1][0] = level;
input[i-1] = new double[9];
input[i-1][0] = spddiff;
input[i-1][1] = voldiff;
input[i-1][2] = occdiff;
input[i-1][3] = u_spddifft;
input[i-1][4] = u_voldifft;
input[i-1][5] = u_occdifft;
input[i-1][6] = d_spddifft;
input[i-1][7] = d_voldifft;
input[i-1][8] = d_occdifft;
Console.WriteLine(spddiff + "," + voldiff + "," + occdiff + "," + u_spddifft + "," + u_voldifft + "," + u_occdifft + "," + d_spddifft + "," + d_voldifft + "," + d_occdifft+","+level);
}
ActivationNetwork network = new ActivationNetwork(
new BipolarSigmoidFunction(1.5),9, 25, 1 );
// create teacher
BackPropagationLearning teacher = new BackPropagationLearning(network);
teacher.Momentum = 0.1;
teacher.LearningRate = 0.01;
// loop
double err=100;
int cnt = 0;
while (err / tdatas.Length >0.00079)
{
// run epoch of learning procedure
err = teacher.RunEpoch( input, output );
Console.WriteLine(err / tdatas.Length);
cnt++;
}
for (int i = 0; i < tdatas.Length-1; i++)
{
if (System.Convert.ToInt32(output[i][0]) != System.Convert.ToInt32(network.Compute(input[i])[0]))
Console.WriteLine("fail");
// Console.WriteLine("chreck:"+Math.Abs((output[i][0] - network.Compute(input[i])[0]))>0,5?:"fail","");
}
}
public EstimationResult Estimate(IEnumerable<IDateValue> dateValues)
{
var data = dateValues.ToArray();
var samplesCount = data.Length - LayerWidth;
var factor = 1.7 / data.Length;
var yMin = data.Min(x => x.Value);
var input = new double[samplesCount][];
var output = new double[samplesCount][];
for (var i = 0; i < samplesCount; i++)
{
input[i] = new double[LayerWidth];
output[i] = new double[1];
for (var j = 0; j < LayerWidth; j++)
input[i][j] = (data[i + j].Value - yMin) * factor - 0.85;
output[i][0] = (data[i + LayerWidth].Value - yMin) * factor - 0.85;
}
var network = new ActivationNetwork(
new BipolarSigmoidFunction(SigmoidAlphaValue),
LayerWidth, LayerWidth * 2, 1);
var teacher = new BackPropagationLearning(network)
{
LearningRate = LearningRate,
Momentum = Momentum
};
var solutionSize = data.Length - LayerWidth;
var solution = new double[solutionSize, 2];
var networkInput = new double[LayerWidth];
for (var j = 0; j < solutionSize; j++)
solution[j, 0] = j + LayerWidth;
TimesLoop.Do(Iterations, () =>
{
teacher.RunEpoch(input, output);
for (int i = 0, n = data.Length - LayerWidth; i < n; i++)
{
for (var j = 0; j < LayerWidth; j++)
networkInput[j] = (data[i + j].Value - yMin) * factor - 0.85;
solution[i, 1] = (network.Compute(networkInput)[0] + 0.85) / factor + yMin;
}
});
return EstimationResult.Create(solution[0, 1], this);
}
static void Main(string[] args)
{
Console.WriteLine("This is a demo application that combines Linear Discriminant Analysis (LDA) and Multilayer Perceptron(MLP).");
double[,] inputs =
{
{ 4, 1 },
{ 2, 4 },
{ 2, 3 },
{ 3, 6 },
{ 4, 4 },
{ 9, 10 },
{ 6, 8 },
{ 9, 5 },
{ 8, 7 },
{ 10, 8 }
};
int[] output =
{
1, 1, 2, 1, 1, 2, 2, 2, 1, 2
};
Console.WriteLine("\r\nProcessing sample data, pease wait...");
//1.1
var lda = new LinearDiscriminantAnalysis(inputs, output);
//1.2 Compute the analysis
lda.Compute();
//1.3
double[,] projection = lda.Transform(inputs);
//both LDA and MLP have a little bit different inputs
//e.x double[,] to double[][], etc.
//e.x. LDA needs int classes and MLP needs classes to be in the range [0..1]
#region convertions
int vector_count = projection.GetLength(0);
int dimensions = projection.GetLength(1);
//====================================================================
// conver for NN
double[][] input2 = new double[vector_count][];
double[][] output2 = new double[vector_count][];
for (int i = 0; i < input2.Length; i++)
{
input2[i] = new double[projection.GetLength(1)];
for (int j = 0; j < projection.GetLength(1); j++)
{
input2[i][j] = projection[i, j];
}
output2[i] = new double[1];
//we turn classes from ints to doubles in the range [0..1], because we use sigmoid for the NN
output2[i][0] = Convert.ToDouble(output[i]) / 10;
}
#endregion
//2.1 create neural network
ActivationNetwork network = new ActivationNetwork(
new SigmoidFunction(2),
dimensions, // inputs neurons in the network
dimensions, // neurons in the first layer
1); // one neuron in the second layer
//2.2 create teacher
BackPropagationLearning teacher = new BackPropagationLearning(network);
//2.3 loop
int p = 0;
while (true)
{
// run epoch of learning procedure
double error = teacher.RunEpoch(input2, output2);
p++;
if (p > 1000000) break;
// instead of iterations we can check error values to see if we need to stop
}
//====================================================================
//3. Classify
double[,] sample = { { 10, 8 } };
double[,] projectedSample = lda.Transform(sample);
double[] projectedSample2 = new double[2];
projectedSample2[0] = projectedSample[0, 0];
projectedSample2[1] = projectedSample[0, 1];
double[] classs = network.Compute(projectedSample2);
Console.WriteLine("========================");
//we convert back to int classes by first rounding and then multipling by 10 (because we devided to 10 before)
//if you do not get the expected result
//- rounding might be a problem
//- try more training
Console.WriteLine(Math.Round(classs[0], 1, MidpointRounding.AwayFromZero)*10);
Console.ReadLine();
}
public void Learn()
{
var network = new ActivationNetwork(new BipolarSigmoidFunction(), Constants.StoneCount, 1);
var teacher = new BackPropagationLearning(network);//new PerceptronLearning(network);
var data = LoadData("4-6-2012-04-24.know");
double error = 1.0;
int index = 0;
while (error > 0.001 && index < 100000) {
error = teacher.RunEpoch(data.Item1, data.Item2);
index++;
}
network.Save("4-6-2012-04-24.bp.net");
var text = "□○○○●○○□○●●□□●□□";
var i = ToDouble(text);//-2
var o = network.Compute(i);
var eval = o[0] * 2 * Constants.StoneCount - Constants.StoneCount;
Console.WriteLine("{0} {1}", text, eval);
}
// Worker thread
void SearchSolution()
{
// initialize input and output values
double[][] input = null;
double[][] output = null;
if (sigmoidType == 0)
{
// unipolar data
input = new double[4][] {
new double[] {0, 0},
new double[] {0, 1},
new double[] {1, 0},
new double[] {1, 1}
};
output = new double[4][] {
new double[] {0},
new double[] {1},
new double[] {1},
new double[] {0}
};
}
else
{
// biipolar data
input = new double[4][] {
new double[] {-1, -1},
new double[] {-1, 1},
new double[] { 1, -1},
new double[] { 1, 1}
};
output = new double[4][] {
new double[] {-1},
new double[] { 1},
new double[] { 1},
new double[] {-1}
};
}
// create perceptron
ActivationNetwork network = new ActivationNetwork(
(sigmoidType == 0) ?
(IActivationFunction)new SigmoidFunction(sigmoidAlphaValue) :
(IActivationFunction)new BipolarSigmoidFunction(sigmoidAlphaValue),
2, 2, 1);
// create teacher
BackPropagationLearning teacher = new BackPropagationLearning(network);
// set learning rate and momentum
teacher.LearningRate = learningRate;
teacher.Momentum = momentum;
// iterations
int iteration = 1;
// statistic files
StreamWriter errorsFile = null;
try
{
// check if we need to save statistics to files
if (saveStatisticsToFiles)
{
// open files
errorsFile = File.CreateText("errors.csv");
}
// erros list
List<double> errorsList = new List<double>();
// loop
while (!needToStop)
{
// run epoch of learning procedure
double error = teacher.RunEpoch(input, output);
errorsList.Add(error);
// save current error
if (errorsFile != null)
{
errorsFile.WriteLine(error);
}
// show current iteration & error
UpdateTextbox(currentIterationBox, iteration.ToString());
//currentIterationBox.Text = iteration.ToString();
UpdateTextbox(currentErrorBox, error.ToString());
//currentErrorBox.Text = error.ToString();
iteration++;
// check if we need to stop
if (error <= learningErrorLimit)
break;
}
// show error's dynamics
double[,] errors = new double[errorsList.Count, 2];
for (int i = 0, n = errorsList.Count; i < n; i++)
{
errors[i, 0] = i;
errors[i, 1] = errorsList[i];
}
errorChart.RangeX = new DoubleRange(0, errorsList.Count - 1);
errorChart.UpdateDataSeries("error", errors);
}
catch (IOException)
{
MessageBox.Show("Failed writing file", "Error", MessageBoxButtons.OK, MessageBoxIcon.Error);
}
finally
{
// close files
if (errorsFile != null)
errorsFile.Close();
}
// enable settings controls
EnableControls(true);
}
private void Button_Click_1(object sender, RoutedEventArgs e)
{
var learningTask = new Task(delegate
{
var learningData = GenerateInputOutput().Take(10);
BackPropagationLearning trainer = new BackPropagationLearning(net);
double prErr = 10000000;
// Ошибка сети
double error = 100;
// Сначала скорость обучения должна быть высока
trainer.LearningRate = 100;
int iteration = 0;
// Обучаем сеть пока ошибка сети станет небольшой
while (error > 0.001)
{
iteration++;
// Получаем ошибку сети
var che = false;
Dispatcher.Invoke(DispatcherPriority.Normal,
new Action(
delegate
{ if (ShowResultsCheckbox.IsChecked != null) che = ShowResultsCheckbox.IsChecked.Value; }));
if (!che)
{
error = trainer.RunEpoch(learningData.Select(i => i.Item1).ToArray(), learningData.Select(i => i.Item2).ToArray());
Dispatcher.Invoke(DispatcherPriority.Normal,
new Action(delegate { label.Content = error + "\nIteration: " + iteration; }));
}
else
{
Dispatcher.Invoke(DispatcherPriority.Normal,
new Action(delegate { label.Content = "See?"; }));
foreach (var data in learningData)
{
Dispatcher.Invoke(DispatcherPriority.Normal,
new Action(
delegate
{ if (ShowResultsCheckbox.IsChecked != null) che = ShowResultsCheckbox.IsChecked.Value; }));
if (!che)
break;
// Если ошибка сети изменилась на небольшое значения, в сравнении ошибкой предыдущей эпохи
Dispatcher.Invoke(DispatcherPriority.Normal, new Action(delegate
{
_answerLable.Content = String.Join(", ", data.Item2.Select(d => d.ToString("N2")));
_learnedLable.Content = String.Join(", ", net.Compute(data.Item1).Select(d => d.ToString("N2")));
im.Source = BitmapSourceConvert.ToBitmapSource(data.Item3);
im.UpdateLayout();
}));
Thread.Sleep(1000);
}
}
if (Math.Abs(error - prErr) < 0.000000001)
{
// Уменьшаем коэффициент скорости обучения на 2
trainer.LearningRate /= 2;
if (trainer.LearningRate < 0.001)
trainer.LearningRate = 0.001;
}
prErr = error;
learningData = GenerateInputOutput().Take(10);
}
});
learningTask.Start();
}
// Worker thread
void SearchSolution( )
{
// number of learning samples
int samples = data.Length - predictionSize - windowSize;
// data transformation factor
double factor = 1.7 / chart.RangeY.Length;
double yMin = chart.RangeY.Min;
// prepare learning data
double[][] input = new double[samples][];
double[][] output = new double[samples][];
for ( int i = 0; i < samples; i++ )
{
input[i] = new double[windowSize];
output[i] = new double[1];
// set input
for ( int j = 0; j < windowSize; j++ )
{
input[i][j] = ( data[i + j] - yMin ) * factor - 0.85;
}
// set output
output[i][0] = ( data[i + windowSize] - yMin ) * factor - 0.85;
}
// create multi-layer neural network
ActivationNetwork network = new ActivationNetwork(
new BipolarSigmoidFunction( sigmoidAlphaValue ),
windowSize, windowSize * 2, 1 );
// create teacher
BackPropagationLearning teacher = new BackPropagationLearning( network );
// set learning rate and momentum
teacher.LearningRate = learningRate;
teacher.Momentum = momentum;
// iterations
int iteration = 1;
// solution array
int solutionSize = data.Length - windowSize;
double[,] solution = new double[solutionSize, 2];
double[] networkInput = new double[windowSize];
// calculate X values to be used with solution function
for ( int j = 0; j < solutionSize; j++ )
{
solution[j, 0] = j + windowSize;
}
// loop
while ( !needToStop )
{
// run epoch of learning procedure
double error = teacher.RunEpoch( input, output ) / samples;
// calculate solution and learning and prediction errors
double learningError = 0.0;
double predictionError = 0.0;
// go through all the data
for ( int i = 0, n = data.Length - windowSize; i < n; i++ )
{
// put values from current window as network's input
for ( int j = 0; j < windowSize; j++ )
{
networkInput[j] = ( data[i + j] - yMin ) * factor - 0.85;
}
// evalue the function
solution[i, 1] = ( network.Compute( networkInput)[0] + 0.85 ) / factor + yMin;
// calculate prediction error
if ( i >= n - predictionSize )
{
predictionError += Math.Abs( solution[i, 1] - data[windowSize + i] );
}
else
{
learningError += Math.Abs( solution[i, 1] - data[windowSize + i] );
}
}
// update solution on the chart
chart.UpdateDataSeries( "solution", solution );
// set current iteration's info
SetText( currentIterationBox, iteration.ToString( ) );
SetText( currentLearningErrorBox, learningError.ToString( "F3" ) );
SetText( currentPredictionErrorBox, predictionError.ToString( "F3" ) );
// increase current iteration
iteration++;
// check if we need to stop
if ( ( iterations != 0 ) && ( iteration > iterations ) )
break;
}
// show new solution
for ( int j = windowSize, k = 0, n = data.Length; j < n; j++, k++ )
{
AddSubItem( dataList, j, solution[k, 1].ToString( ) );
}
// enable settings controls
EnableControls( true );
}
private void DoWork(object sender, DoWorkEventArgs e)
{
BackgroundWorker worker = sender as BackgroundWorker;
e.Result = new double[0];
if (worker.CancellationPending)
{
e.Cancel = true;
return;
}
TrainingDialog obj = (TrainingDialog)e.Argument;
_Series++;
#region Prepare data to be trained. Involves copying.
int numTrainingSets = obj._DataSet.Training.Rows.Count;
ArrayList inputs = new ArrayList();
ArrayList outputs = new ArrayList();
for (int i = 0; i < numTrainingSets; i++)
{
// Input data
double[] inData = new double[_NumInputNeurons];
inData[0] = obj._DataSet.Training[i].LipCornerLeftX;
inData[1] = obj._DataSet.Training[i].LipCornerLeftY;
inData[2] = obj._DataSet.Training[i].LipCornerRightX;
inData[3] = obj._DataSet.Training[i].LipCornerRightY;
inData[4] = obj._DataSet.Training[i].LipUpLeftX;
inData[5] = obj._DataSet.Training[i].LipUpLeftY;
inData[6] = obj._DataSet.Training[i].LipUpCenterX;
inData[7] = obj._DataSet.Training[i].LipUpCenterY;
inData[8] = obj._DataSet.Training[i].LipUpRightX;
inData[9] = obj._DataSet.Training[i].LipUpRightY;
inData[10] = obj._DataSet.Training[i].LipBottomLeftX;
inData[11] = obj._DataSet.Training[i].LipBottomLeftY;
inData[12] = obj._DataSet.Training[i].LipBottomCenterX;
inData[13] = obj._DataSet.Training[i].LipBottomCenterY;
inData[14] = obj._DataSet.Training[i].LipBottomRightX;
inData[15] = obj._DataSet.Training[i].LipBottomRightY;
inData[16] = obj._DataSet.Training[i].LeftEyeCenterX;
inData[17] = obj._DataSet.Training[i].LeftEyeCenterY;
inData[18] = obj._DataSet.Training[i].LeftLidBottomX;
inData[19] = obj._DataSet.Training[i].LeftLidBottomY;
inData[20] = obj._DataSet.Training[i].LeftLidCornerLeftX;
inData[21] = obj._DataSet.Training[i].LeftLidCornerLeftY;
inData[22] = obj._DataSet.Training[i].LeftLidCornerRightX;
inData[23] = obj._DataSet.Training[i].LeftLidCornerRightY;
inData[24] = obj._DataSet.Training[i].LeftLidUpX;
inData[25] = obj._DataSet.Training[i].LeftLidUpY;
inData[26] = obj._DataSet.Training[i].MouthCenterX;
inData[27] = obj._DataSet.Training[i].MouthCenterY;
inData[28] = obj._DataSet.Training[i].RightEyeCenterX;
inData[29] = obj._DataSet.Training[i].RightEyeCenterY;
inData[30] = obj._DataSet.Training[i].RightLidBottomX;
inData[31] = obj._DataSet.Training[i].RightLidBottomY;
inData[32] = obj._DataSet.Training[i].RightLidCornerLeftX;
inData[33] = obj._DataSet.Training[i].RightLidCornerLeftY;
inData[34] = obj._DataSet.Training[i].RightLidCornerRightX;
inData[35] = obj._DataSet.Training[i].RightLidCornerRightY;
inData[36] = obj._DataSet.Training[i].RightLidUpX;
inData[37] = obj._DataSet.Training[i].RightLidUpY;
inputs.Add(inData);
// Output data
double[] outData = new double[_NumOutputNeurons];
int eid = obj._DataSet.Training[i].ExpressionOID;
string expression = obj._DataSet.Expression.FindByExpressionOID(eid).Expression.ToLower();
outData[0] = expression.Contains("anger") ? 1 : 0;
outData[1] = expression.Contains("disg") ? 1 : 0;
outData[2] = expression.Contains("fear") ? 1 : 0;
outData[3] = expression.Contains("happy") ? 1 : 0;
outData[4] = expression.Contains("neutr") ? 1 : 0;
outData[5] = expression.Contains("sad") ? 1 : 0;
outData[6] = expression.Contains("surp") ? 1 : 0;
outputs.Add(outData);
}
#endregion
#region Norm datasets per input neuron
for (int j = 0; j < _NumInputNeurons; j++)
{
double min = 100000000.0;
double max = -100000000.0;
for (int i = 0; i < numTrainingSets; i++)
{
double cur = ((double[])inputs[i])[j];
if (min > cur)
{
min = cur;
}
if (max < cur)
{
max = cur;
}
}
for (int i = 0; (max - min) != 0 && i < numTrainingSets; i++)
{
((double[])inputs[i])[j] = (((double[])inputs[i])[j] - min) / (max - min);
}
}
#endregion
#region Pick random train-, validate and test datasets
// Like Mr. Schneider ;)
int numTestDataSets = (int)Math.Floor((double)numTrainingSets * 0.1);
int numValidationDataSets = (int)Math.Floor((double)(numTrainingSets - numTestDataSets) * 0.2);
int numTrainDataSets = numTrainingSets - numTestDataSets - numValidationDataSets;
Random rand = new Random();
// Get random training data
double[][] trainingInputs = new double[numTrainDataSets][];
double[][] trainingOutputs = new double[numTrainDataSets][];
for (int i = 0; i < trainingInputs.GetLength(0); i++)
{
int idx = rand.Next(inputs.Count);
trainingInputs[i] = (double[])inputs[idx];
trainingOutputs[i] = (double[])outputs[idx];
inputs.Remove(idx);
outputs.Remove(idx);
}
// Get random validation data
double[][] validateInputs = new double[numValidationDataSets][];
double[][] validateOutputs = new double[numValidationDataSets][];
for (int i = 0; i < validateInputs.GetLength(0); i++)
{
int idx = rand.Next(numTrainDataSets);
validateInputs[i] = trainingInputs[idx];
validateOutputs[i] = trainingOutputs[idx];
}
// Get random test data
double[][] testInputs = new double[numTestDataSets][];
double[][] testOutputs = new double[numTestDataSets][];
for (int i = 0; i < testInputs.GetLength(0); i++)
{
int idx = rand.Next(inputs.Count);
testInputs[i] = (double[])inputs[idx];
testOutputs[i] = (double[])outputs[idx];
inputs.Remove(idx);
outputs.Remove(idx);
}
#endregion
#region
_Network.SetActivationFunction(new SigmoidFunction(_SigmoidAlpha));
BackPropagationLearning teacher = new BackPropagationLearning(_Network);
teacher.LearningRate = obj._LearningRate;
teacher.Momentum = obj._Momentum;
double error = 1;
double maxError = -10000.0;
int maxIterations = obj._MaxIterations;
double epsilon = obj._Epsilon;
// Prepare the error Chart
double[,] errorValues = new double[maxIterations, 2];
ProgressState state = new ProgressState();
double[] errors = new double[maxIterations];
for (int i = 0; i < maxIterations || error <= epsilon; i++)
{
if (maxIterations > 0)
{
if (i >= maxIterations)
{
break;
}
}
if (epsilon > 0.0)
{
if (error <= epsilon)
{
break;
}
}
// Abort if use requested it
if (worker.CancellationPending)
{
e.Cancel = true;
break;
}
// Train
error = teacher.RunEpoch(trainingInputs, trainingOutputs);
// Store error for result
errors[i] = error;
state.Error = error;
// Plot size
if (maxError < error)
{
maxError = error;
}
// Plot values
errorValues[i, 0] = i;
errorValues[i, 1] = error;
// Report progress
if (i % 10 == 0 || i == (maxIterations - 1))
{
state.Iteration = i + 1;
state.Error = error;
worker.ReportProgress((int)((float)(100 * (i + 1)) / (float)maxIterations), state);
}
}
e.Result = errors;
obj._Chart.RangeX = new DoubleRange(0.0, (double)maxIterations);
obj._Chart.RangeY = new DoubleRange(0.0, maxError);
// add new data series to the chart
obj._Chart.AddDataSeries("Error " + _Series, System.Drawing.Color.DarkGreen, Chart.SeriesType.ConnectedDots, 2);
// update the chart
obj._Chart.UpdateDataSeries("Error " + _Series, errorValues);
#endregion
}
