public void LoadData_Check()
{
// Arrange
ANeuralNetwork ANN = new ANeuralNetwork();
List <List <double> > ExpectedOutput = new List <List <double> > {
new List <double> {
1, 2, 3
}, new List <double> {
4, 5, 6
}
};
// Act
List <List <double> > Output = ANN.Load_Data("TestDataInput.dat");
// Asert
bool ItemsMatch = true;
for (int row = 0; row < Output.Count; row++)
{
for (int column = 0; column < Output[row].Count; column++)
{
Console.Write(Output[row][column] + " ");
if (ExpectedOutput[row][column] != Output[row][column])
{
ItemsMatch = false;
}
}
Console.WriteLine();
}
Assert.IsTrue(ItemsMatch);
}
public void CheckOne()
{
ANeuralNetwork network = new ANeuralNetwork();
network.Load("..\\..\\..\\..\\savedData.txt");
var sampleDirs = Directory.GetDirectories("..\\..\\..\\..\\Data\\TestData\\").ToList();
int fCount = Directory.GetFiles("..\\..\\..\\..\\Data\\GroundData\\", "*.png", SearchOption.AllDirectories).Length;
var poliManager = new PolynomialManager();
poliManager.InitBasis(15, 100);
double count = 0;
foreach (var dir in sampleDirs)
{
var files = Directory.GetFiles(dir, "*.png", SearchOption.TopDirectoryOnly).ToList();
var index = new Random().Next(files.Count);
var file = files[index];
var value = Convert.ToInt32(new StreamReader(dir + "\\" + "value.txt").ReadToEnd());
++count;
ComplexMoments tmpMoments;
ProcessOneImage(file, poliManager, 100, out tmpMoments);
CvInvoke.Imshow(dir, tmpMoments.Real);
var tmpInput = tmpMoments.ToListOfDouble();
var output = network.Predict(tmpInput);
var predictedValue = Convert.ToInt32(output.IndexOf(output.Max()));
Console.WriteLine(predictedValue);
CvInvoke.WaitKey();
}
}
public void TrainMiniBatch_ConvergenceTest_ValuesNOT01()
{
// Arrange
ANeuralNetwork ANN = new ANeuralNetwork();
ANN.Load_Network("Test.net");
List <List <double> > InputData = new List <List <double> > {
new List <double> {
10, 10
}
};
List <List <double> > OutputData = new List <List <double> > {
new List <double> {
10
}
};
// Act
ANN.Train_MiniBatch(InputData, OutputData, InputData.Count, "MiniBatch_Convergence.dat");
List <double> Result = ANN.Get_Network_Output(InputData[0]);
// Assert
Console.Write("Output = " + Result[0]);
Assert.AreNotEqual(OutputData[0][0], Math.Round(Result[0], 1));
}
public void SplitData_SampleSizeLargerThanDataSize()
{
// Arrange
ANeuralNetwork ANN = new ANeuralNetwork();
List <List <double> > TestData = new List <List <double> > {
new List <double> {
1, 2
}, new List <double> {
3, 4
}, new List <double> {
5, 6
}, new List <double> {
7, 8
}
};
Exception expectedException = null;
// Act
try
{
ANN.SplitDataSet(TestData, 5);
}
catch (Exception ex)
{
expectedException = ex;
}
// Assert
Assert.IsNotNull(expectedException);
}
void CreatePopulation(int count)
{
List <uint> Configurate = new List <uint>
{
2,
2,
1
};
Nr = new List <NeuralIndividual>();
inputs = new List <List <double> >();
outputs = new List <List <double> >();
for (int i = 0; i < count; i++)
{
inputs.Clear();
outputs.Clear();
var tmpNr = new ANeuralNetwork(Configurate, ActivationType.PositiveSygmoid, 1);
tmpNr.LoadData("..\\..\\..\\..\\PerceptronData.txt", inputs, outputs);
tmpNr.RandomInit();
//var tmpError = tmpNr.BackPropTraining(inputs, outputs, 1, 0.0001, 1, false);//Обучение сети метдом оьбратного распространения ошибки
tmpNr.IsTrained = true;
var tm = CheckAccuracy(tmpNr);
NeuralIndividual tmpNi = new NeuralIndividual(tmpNr, tm);
Nr.Add(tmpNi);
}
}
public void TrainMiniBatch_XORGateTest_SingleBatch()
{
// Arrange
ANeuralNetwork ANN = new ANeuralNetwork();
ANN.Create_Network(new List <int> {
2, 3, 1
});
List <List <double> > InputData = new List <List <double> > {
new List <double> {
1, 1
}, new List <double> {
0, 1
}, new List <double> {
1, 0
}, new List <double> {
0, 0
},
};
List <List <double> > OutputData = new List <List <double> > {
new List <double> {
0
}, new List <double> {
1
}, new List <double> {
1
}, new List <double> {
0
}
};
List <List <double> > ValidationData = new List <List <double> > {
new List <double> {
1, 1
}, new List <double> {
0, 1
}, new List <double> {
1, 0
}, new List <double> {
0, 0
},
};
List <double> ExpectedOutput = new List <double> {
0, 1, 1, 0
};
// Act
ANN.Train_MiniBatch(InputData, OutputData, InputData.Count, "XOR_SinleBatch.dat");
List <double> Results = new List <double> {
};
for (int i = 0; i < 4; i++)
{
List <double> result = ANN.Get_Network_Output(ValidationData[i]);
Results.Add(Math.Round(result[0], 0));
Console.Write("Output = " + Results[i] + " ");
}
// Assert
CollectionAssert.AreEqual(ExpectedOutput, Results);
}
string Recognize(ComplexMoments moments)
{
ANeuralNetwork networt = new ANeuralNetwork();
var output = networt.Predict(moments.ToListOfDouble());
return(output.IndexOf(output.Max()).ToString());
}
public void Train(double accuracy, int populationCount)
{
Stopwatch stopwatch = new Stopwatch();
double curError = 1;
CreatePopulation(populationCount);
int iter = 0;
while (curError > accuracy)
{
stopwatch.Start();
CrossPopulation();
curError = CheckAccuracy();
if (iter % 100 == 0)
{
stopwatch.Stop();
var ts = stopwatch.Elapsed;
stopwatch.Reset();
Console.WriteLine("Iteration: " + iter + "\tError: " + curError + "\tTime: " + ts.TotalSeconds);
}
iter++;
}
result = Nr[0].network;
}
public void SplitData()
{
// Arrange
ANeuralNetwork ANN = new ANeuralNetwork();
List <List <double> > TestData = new List <List <double> > {
new List <double> {
1, 2, 3, 4
}, new List <double> {
5, 6, 7, 8
}
};
List <List <List <double> > > ExpectedOutput = new List <List <List <double> > > {
new List <List <double> > {
new List <double> {
1, 2
}
}, new List <List <double> > {
new List <double> {
3, 4
}
}, new List <List <double> > {
new List <double> {
5, 6
}
}, new List <List <double> > {
new List <double> {
7, 8
}
}
};
// Act
ANN.Create_Network(new List <int> {
2, 1, 2
});
List <List <List <double> > > Output = ANN.SplitDataSet(TestData, 1);
// Assert
bool ItemsMatch = true;
for (int ItemInOutput = 0; ItemInOutput < Output.Count; ItemInOutput++)
{
for (int row = 0; row < Output[ItemInOutput].Count; row++)
{
for (int column = 0; column < Output[ItemInOutput][row].Count; column++)
{
Console.Write(Output[ItemInOutput][row][column] + " ");
if (ExpectedOutput[ItemInOutput][row][column] != Output[ItemInOutput][row][column])
{
ItemsMatch = false;
}
}
Console.WriteLine();
}
Console.WriteLine();
}
Assert.IsTrue(ItemsMatch);
}
double CheckAccuracy(ANeuralNetwork ann)
{
double error = 0;
for (int i = 0; i < inputs.Count; i++)
{
for (int j = 0; j < outputs[i].Count; j++)
{
var tmp = ann.Predict(inputs[i])[j];
error += (outputs[i][j] - ann.Predict(inputs[i])[j]) * (outputs[i][j] - ann.Predict(inputs[i])[j]);
}
}
return(Math.Sqrt(error / inputs.Count));
}
public bool Train(List <uint> layers, SortedDictionary <string, List <ComplexMoments> > moments, int maxIters = 100000, double eps = 0.1, double speed = 0.1)
{
ANeuralNetwork network = new ANeuralNetwork(layers, ANeuralNetwork.ActivationType.BipolarSygmoid, 1);
var inputs = new List <List <double> >();
var outputs = new List <List <double> >();
var max = int.MinValue;
foreach (var value in moments)
{
if (value.Value.Count > max)
{
max = value.Value.Count;
}
}
int iter = 0;
int counter = 0;
while (iter < max && counter < layers[^ 1])
{
for (var i = 0; i < moments.Keys.Count; ++i)
{
var output = new List <double>();
for (var j = 0; j < moments.Keys.Count; j++)
{
output.Add(j == i ? 1 : 0);
}
counter++;
if (moments[i.ToString()].Count - 1 <= iter)
{
continue;
}
var input = moments[i.ToString()][iter].ToListOfDouble();
inputs.Add(input);
outputs.Add(output);
counter--;
iter++;
}
}
Console.WriteLine("Данные на входе, начинаем обучение");
network.BackPropTraining(inputs, outputs, maxIters, eps, speed, true, 1);
network.Save("..\\..\\..\\..\\savedData.txt");
return(true);
}
public void GetNetworkOutput_CheckOutputCorrect()
{
// Arrange
ANeuralNetwork ANN = new ANeuralNetwork();
ANN.Load_Network("Test.net");
List <double> InputData = new List <double> {
0.1, 0.9
};
double DesiredOutput = 0.60761339;
// Act
List <double> Output = ANN.Get_Network_Output(InputData);
// Assert
Assert.AreEqual(DesiredOutput, Math.Round(Output[0], 8));
}
static void Main(string[] args)
{
ANeuralNetwork tmp = new ANeuralNetwork();
tmp.Load("..\\..\\..\\..\\PerceptronSavedData.txt");
var line = Console.ReadLine();
while (line!="q")
{
var tmpInput = line.Split().Select(double.Parse).ToList();
var tmpOutput = tmp.Predict(tmpInput);
foreach (var element in tmpOutput)
{
Console.WriteLine(element);
}
line = Console.ReadLine();
}
}
static void BackPropagation()
{
List <uint> Configurate = new List <uint>(); //Создаем конфигурацию сети, состоящую из трех слоев
Configurate.Add(2); //1-ой входной слой сети с двумя нейронами
//2-ой слой сети с двумя нейронами
Configurate.Add(5);
Configurate.Add(3);
Configurate.Add(1);//4-й выходнрой слой сети с одним нейроном
ANeuralNetwork NR = new ANeuralNetwork(Configurate, ANeuralNetwork.ActivationType.PositiveSygmoid, 1);
List <List <double> > inputs = new List <List <double> >();
List <List <double> > outputs = new List <List <double> >(); //создаем пустые вектора под входные и выходные данные
NR.LoadData("..\\..\\..\\..\\PerceptronData.txt", inputs, outputs); // выгружаем данные из файла
// инцилизируем веса сети случайным образом
NR.BackPropTraining(inputs, outputs, 3000000, 0.01, 1, true); //Обучение сети метдом оьбратного распространения ошибки
Console.WriteLine(NR.GetType()); //в каждые 100 итераций выводит данные в строку
NR.Save("..\\..\\..\\..\\PerceptronSavedData.txt"); //сохраняем в файл
}
public void LoadInputFile_InvaildNetworkDescription()
{
// Arrange
string Filename = "Test_InvalidNetworkDescription.net";
ANeuralNetwork ANN = new ANeuralNetwork();
Exception expectedException = null;
// Act
try
{
ANN.Load_Network(Filename);
}
catch (Exception ex)
{
expectedException = ex;
}
// Assert
Assert.IsNotNull(expectedException);
}
public void LoadInputFile_FileNotExists()
{
// Arrange
string Filename = "DoesNotExist.net";
ANeuralNetwork ANN = new ANeuralNetwork();
Exception expectedException = null;
// Act
try
{
ANN.Load_Network(Filename);
}
catch (Exception ex)
{
expectedException = ex;
}
// Assert
Assert.IsNotNull(expectedException);
}
public void LoadData_FileExists()
{
// Arrange
string Filename = "TestDataInput.dat";
ANeuralNetwork ANN = new ANeuralNetwork();
Exception expectedException = null;
// Act
try
{
ANN.Load_Data(Filename);
}
catch (Exception ex)
{
expectedException = ex;
}
// Assert
Assert.IsNull(expectedException);
}
public void Check()
{
ANeuralNetwork network = new ANeuralNetwork();
network.Load("..\\..\\..\\..\\savedData.txt");
var sampleDirs = Directory.GetDirectories("..\\..\\..\\..\\Data\\TestData\\").ToList();
int fCount = Directory.GetFiles("..\\..\\..\\..\\Data\\GroundData\\", "*.png", SearchOption.AllDirectories).Length;
double precision = 0;
var poliManager = new PolynomialManager();
poliManager.InitBasis(15, 100);
double count = 0;
double trueCount = 0;
foreach (var dir in sampleDirs)
{
var files = Directory.GetFiles(dir, "*.png", SearchOption.TopDirectoryOnly).ToList();
var value = Convert.ToInt32(new StreamReader(dir + "\\" + "value.txt").ReadToEnd());
foreach (var file in files)
{
++count;
ComplexMoments tmpMoments;
ProcessOneImage(file, poliManager, 100, out tmpMoments);
var tmpInput = tmpMoments.ToListOfDouble();
var output = network.Predict(tmpInput);
var predictedValue = Convert.ToInt32(output.IndexOf(output.Max()));
if (predictedValue == value)
{
//++trueCount;
precision += Convert.ToDouble(100 / (double)fCount);
}
}
Console.WriteLine("Точность " + precision + "%");
//Console.WriteLine("Точность " + trueCount/count + "%");
}
}
public void LoadInputFile_CheckLoadedWeights()
{
// Arrange
string Filename = "Test.net";
ANeuralNetwork ANN = new ANeuralNetwork();
List <double> ExpectedWeights = new List <double> {
0.1, 0.2, 0.1
};
// Act
ANN.Load_Network(Filename);
List <double> Weights = new List <double> {
};
for (int Weight = 0; Weight < ANN.NeuralNetwork[0][0].Weights.Count; Weight++)
{
Weights.Add(ANN.NeuralNetwork[0][0].Weights[Weight]);
}
// Assert
CollectionAssert.AreEqual(ExpectedWeights, Weights);
}
public void SaveNetwork()
{
// Arrange
ANeuralNetwork ANN = new ANeuralNetwork();
ANN.Create_Network(new List <int> {
2, 3, 1
});
Exception expectedException = null;
// Act
try
{
ANN.Save_Network("TestSave.net");
}
catch (Exception ex)
{
expectedException = ex;
}
// Assert
Assert.IsNull(expectedException);
}
public NeuralIndividual(ANeuralNetwork tmpNr, double tmpError) : this()
{
network = tmpNr;
error = tmpError;
}
void CrossPopulation()
{
List <Tuple <NeuralIndividual, NeuralIndividual> > neuralPair = new List <Tuple <NeuralIndividual, NeuralIndividual> >();
List <uint> Configurate = new List <uint>
{
2,
2,
1
};
ANeuralNetwork ann = new ANeuralNetwork(Configurate, ActivationType.PositiveSygmoid, 1);
ann.RandomInit();
int tmpCount = Nr.Count;
List <int> usedIndex = new List <int>();
for (int i = 0; i < tmpCount; i++)
{
int rand = new Random().Next(tmpCount);
while (usedIndex.Contains(rand))
{
rand = new Random().Next(tmpCount);
}
usedIndex.Add(rand);
if (rand == tmpCount - 1)
{
for (var layerIdx = 0; layerIdx < Nr[i].network.Weights.Count; layerIdx++)
{
for (var fromIdx = 0; fromIdx < Nr[i].network.Weights[layerIdx].Count; fromIdx++)
{
for (var toIdx = 0; toIdx < Nr[i].network.Weights[layerIdx][fromIdx].Count; toIdx++)
{
var tmpRand = new Random().Next(1000);
if (tmpRand >= 50)
{
ann.Weights[layerIdx][fromIdx][toIdx] = Nr[rand].network.Weights[layerIdx][fromIdx][toIdx];
}
else
{
ann.Weights[layerIdx][fromIdx][toIdx] = Nr[0].network.Weights[layerIdx][fromIdx][toIdx];
}
}
}
}
}
else
{
for (var layerIdx = 0; layerIdx < Nr[i].network.Weights.Count; layerIdx++)
{
for (var fromIdx = 0; fromIdx < Nr[i].network.Weights[layerIdx].Count; fromIdx++)
{
for (var toIdx = 0; toIdx < Nr[i].network.Weights[layerIdx][fromIdx].Count; toIdx++)
{
var tmpRand = new Random().Next(1000);
if (tmpRand >= 50)
{
ann.Weights[layerIdx][fromIdx][toIdx] = Nr[rand].network.Weights[layerIdx][fromIdx][toIdx];
}
else
{
ann.Weights[layerIdx][fromIdx][toIdx] = Nr[rand + 1].network.Weights[layerIdx][fromIdx][toIdx];
}
}
}
}
}
ann.IsTrained = true;
var err = CheckAccuracy(ann);
Nr.Add(new NeuralIndividual(ann, err));
}
for (int i = 0; i < Nr.Count - 1; i++)
{
for (int j = 0; j < Nr.Count - i - 1; j++)
{
if (Nr[j].error > Nr[j + 1].error)
{
var temp = Nr[j];
Nr[j] = Nr[j + 1];
Nr[j + 1] = temp;
}
}
}
Nr.RemoveRange(tmpCount, tmpCount);
}
public void IrisClasificationTest()
{
#region Training Data
//Encoded
// Iris setosa = 1 0 0
// Iris versicolor = 0 1 0
// Iris virginica = 0 0 1
string[] rawData = new[]
{
"setosa",
"versicolor",
"virginica"
};
double[][] encodedData = new double[150][];
encodedData[0] = new double[] { 5.1, 3.5, 1.4, 0.2, 0, 0, 1 };
encodedData[1] = new double[] { 4.9, 3.0, 1.4, 0.2, 0, 0, 1 };
encodedData[2] = new double[] { 4.7, 3.2, 1.3, 0.2, 0, 0, 1 };
encodedData[3] = new double[] { 4.6, 3.1, 1.5, 0.2, 0, 0, 1 };
encodedData[4] = new double[] { 5.0, 3.6, 1.4, 0.2, 0, 0, 1 };
encodedData[5] = new double[] { 5.4, 3.9, 1.7, 0.4, 0, 0, 1 };
encodedData[6] = new double[] { 4.6, 3.4, 1.4, 0.3, 0, 0, 1 };
encodedData[7] = new double[] { 5.0, 3.4, 1.5, 0.2, 0, 0, 1 };
encodedData[8] = new double[] { 4.4, 2.9, 1.4, 0.2, 0, 0, 1 };
encodedData[9] = new double[] { 4.9, 3.1, 1.5, 0.1, 0, 0, 1 };
encodedData[10] = new double[] { 5.4, 3.7, 1.5, 0.2, 0, 0, 1 };
encodedData[11] = new double[] { 4.8, 3.4, 1.6, 0.2, 0, 0, 1 };
encodedData[12] = new double[] { 4.8, 3.0, 1.4, 0.1, 0, 0, 1 };
encodedData[13] = new double[] { 4.3, 3.0, 1.1, 0.1, 0, 0, 1 };
encodedData[14] = new double[] { 5.8, 4.0, 1.2, 0.2, 0, 0, 1 };
encodedData[15] = new double[] { 5.7, 4.4, 1.5, 0.4, 0, 0, 1 };
encodedData[16] = new double[] { 5.4, 3.9, 1.3, 0.4, 0, 0, 1 };
encodedData[17] = new double[] { 5.1, 3.5, 1.4, 0.3, 0, 0, 1 };
encodedData[18] = new double[] { 5.7, 3.8, 1.7, 0.3, 0, 0, 1 };
encodedData[19] = new double[] { 5.1, 3.8, 1.5, 0.3, 0, 0, 1 };
encodedData[20] = new double[] { 5.4, 3.4, 1.7, 0.2, 0, 0, 1 };
encodedData[21] = new double[] { 5.1, 3.7, 1.5, 0.4, 0, 0, 1 };
encodedData[22] = new double[] { 4.6, 3.6, 1.0, 0.2, 0, 0, 1 };
encodedData[23] = new double[] { 5.1, 3.3, 1.7, 0.5, 0, 0, 1 };
encodedData[24] = new double[] { 4.8, 3.4, 1.9, 0.2, 0, 0, 1 };
encodedData[25] = new double[] { 5.0, 3.0, 1.6, 0.2, 0, 0, 1 };
encodedData[26] = new double[] { 5.0, 3.4, 1.6, 0.4, 0, 0, 1 };
encodedData[27] = new double[] { 5.2, 3.5, 1.5, 0.2, 0, 0, 1 };
encodedData[28] = new double[] { 5.2, 3.4, 1.4, 0.2, 0, 0, 1 };
encodedData[29] = new double[] { 4.7, 3.2, 1.6, 0.2, 0, 0, 1 };
encodedData[30] = new double[] { 4.8, 3.1, 1.6, 0.2, 0, 0, 1 };
encodedData[31] = new double[] { 5.4, 3.4, 1.5, 0.4, 0, 0, 1 };
encodedData[32] = new double[] { 5.2, 4.1, 1.5, 0.1, 0, 0, 1 };
encodedData[33] = new double[] { 5.5, 4.2, 1.4, 0.2, 0, 0, 1 };
encodedData[34] = new double[] { 4.9, 3.1, 1.5, 0.1, 0, 0, 1 };
encodedData[35] = new double[] { 5.0, 3.2, 1.2, 0.2, 0, 0, 1 };
encodedData[36] = new double[] { 5.5, 3.5, 1.3, 0.2, 0, 0, 1 };
encodedData[37] = new double[] { 4.9, 3.1, 1.5, 0.1, 0, 0, 1 };
encodedData[38] = new double[] { 4.4, 3.0, 1.3, 0.2, 0, 0, 1 };
encodedData[39] = new double[] { 5.1, 3.4, 1.5, 0.2, 0, 0, 1 };
encodedData[40] = new double[] { 5.0, 3.5, 1.3, 0.3, 0, 0, 1 };
encodedData[41] = new double[] { 4.5, 2.3, 1.3, 0.3, 0, 0, 1 };
encodedData[42] = new double[] { 4.4, 3.2, 1.3, 0.2, 0, 0, 1 };
encodedData[43] = new double[] { 5.0, 3.5, 1.6, 0.6, 0, 0, 1 };
encodedData[44] = new double[] { 5.1, 3.8, 1.9, 0.4, 0, 0, 1 };
encodedData[45] = new double[] { 4.8, 3.0, 1.4, 0.3, 0, 0, 1 };
encodedData[46] = new double[] { 5.1, 3.8, 1.6, 0.2, 0, 0, 1 };
encodedData[47] = new double[] { 4.6, 3.2, 1.4, 0.2, 0, 0, 1 };
encodedData[48] = new double[] { 5.3, 3.7, 1.5, 0.2, 0, 0, 1 };
encodedData[49] = new double[] { 5.0, 3.3, 1.4, 0.2, 0, 0, 1 };
encodedData[50] = new double[] { 7.0, 3.2, 4.7, 1.4, 0, 1, 0 };
encodedData[51] = new double[] { 6.4, 3.2, 4.5, 1.5, 0, 1, 0 };
encodedData[52] = new double[] { 6.9, 3.1, 4.9, 1.5, 0, 1, 0 };
encodedData[53] = new double[] { 5.5, 2.3, 4.0, 1.3, 0, 1, 0 };
encodedData[54] = new double[] { 6.5, 2.8, 4.6, 1.5, 0, 1, 0 };
encodedData[55] = new double[] { 5.7, 2.8, 4.5, 1.3, 0, 1, 0 };
encodedData[56] = new double[] { 6.3, 3.3, 4.7, 1.6, 0, 1, 0 };
encodedData[57] = new double[] { 4.9, 2.4, 3.3, 1.0, 0, 1, 0 };
encodedData[58] = new double[] { 6.6, 2.9, 4.6, 1.3, 0, 1, 0 };
encodedData[59] = new double[] { 5.2, 2.7, 3.9, 1.4, 0, 1, 0 };
encodedData[60] = new double[] { 5.0, 2.0, 3.5, 1.0, 0, 1, 0 };
encodedData[61] = new double[] { 5.9, 3.0, 4.2, 1.5, 0, 1, 0 };
encodedData[62] = new double[] { 6.0, 2.2, 4.0, 1.0, 0, 1, 0 };
encodedData[63] = new double[] { 6.1, 2.9, 4.7, 1.4, 0, 1, 0 };
encodedData[64] = new double[] { 5.6, 2.9, 3.6, 1.3, 0, 1, 0 };
encodedData[65] = new double[] { 6.7, 3.1, 4.4, 1.4, 0, 1, 0 };
encodedData[66] = new double[] { 5.6, 3.0, 4.5, 1.5, 0, 1, 0 };
encodedData[67] = new double[] { 5.8, 2.7, 4.1, 1.0, 0, 1, 0 };
encodedData[68] = new double[] { 6.2, 2.2, 4.5, 1.5, 0, 1, 0 };
encodedData[69] = new double[] { 5.6, 2.5, 3.9, 1.1, 0, 1, 0 };
encodedData[70] = new double[] { 5.9, 3.2, 4.8, 1.8, 0, 1, 0 };
encodedData[71] = new double[] { 6.1, 2.8, 4.0, 1.3, 0, 1, 0 };
encodedData[72] = new double[] { 6.3, 2.5, 4.9, 1.5, 0, 1, 0 };
encodedData[73] = new double[] { 6.1, 2.8, 4.7, 1.2, 0, 1, 0 };
encodedData[74] = new double[] { 6.4, 2.9, 4.3, 1.3, 0, 1, 0 };
encodedData[75] = new double[] { 6.6, 3.0, 4.4, 1.4, 0, 1, 0 };
encodedData[76] = new double[] { 6.8, 2.8, 4.8, 1.4, 0, 1, 0 };
encodedData[77] = new double[] { 6.7, 3.0, 5.0, 1.7, 0, 1, 0 };
encodedData[78] = new double[] { 6.0, 2.9, 4.5, 1.5, 0, 1, 0 };
encodedData[79] = new double[] { 5.7, 2.6, 3.5, 1.0, 0, 1, 0 };
encodedData[80] = new double[] { 5.5, 2.4, 3.8, 1.1, 0, 1, 0 };
encodedData[81] = new double[] { 5.5, 2.4, 3.7, 1.0, 0, 1, 0 };
encodedData[82] = new double[] { 5.8, 2.7, 3.9, 1.2, 0, 1, 0 };
encodedData[83] = new double[] { 6.0, 2.7, 5.1, 1.6, 0, 1, 0 };
encodedData[84] = new double[] { 5.4, 3.0, 4.5, 1.5, 0, 1, 0 };
encodedData[85] = new double[] { 6.0, 3.4, 4.5, 1.6, 0, 1, 0 };
encodedData[86] = new double[] { 6.7, 3.1, 4.7, 1.5, 0, 1, 0 };
encodedData[87] = new double[] { 6.3, 2.3, 4.4, 1.3, 0, 1, 0 };
encodedData[88] = new double[] { 5.6, 3.0, 4.1, 1.3, 0, 1, 0 };
encodedData[89] = new double[] { 5.5, 2.5, 4.0, 1.3, 0, 1, 0 };
encodedData[90] = new double[] { 5.5, 2.6, 4.4, 1.2, 0, 1, 0 };
encodedData[91] = new double[] { 6.1, 3.0, 4.6, 1.4, 0, 1, 0 };
encodedData[92] = new double[] { 5.8, 2.6, 4.0, 1.2, 0, 1, 0 };
encodedData[93] = new double[] { 5.0, 2.3, 3.3, 1.0, 0, 1, 0 };
encodedData[94] = new double[] { 5.6, 2.7, 4.2, 1.3, 0, 1, 0 };
encodedData[95] = new double[] { 5.7, 3.0, 4.2, 1.2, 0, 1, 0 };
encodedData[96] = new double[] { 5.7, 2.9, 4.2, 1.3, 0, 1, 0 };
encodedData[97] = new double[] { 6.2, 2.9, 4.3, 1.3, 0, 1, 0 };
encodedData[98] = new double[] { 5.1, 2.5, 3.0, 1.1, 0, 1, 0 };
encodedData[99] = new double[] { 5.7, 2.8, 4.1, 1.3, 0, 1, 0 };
encodedData[100] = new double[] { 6.3, 3.3, 6.0, 2.5, 1, 0, 0 };
encodedData[101] = new double[] { 5.8, 2.7, 5.1, 1.9, 1, 0, 0 };
encodedData[102] = new double[] { 7.1, 3.0, 5.9, 2.1, 1, 0, 0 };
encodedData[103] = new double[] { 6.3, 2.9, 5.6, 1.8, 1, 0, 0 };
encodedData[104] = new double[] { 6.5, 3.0, 5.8, 2.2, 1, 0, 0 };
encodedData[105] = new double[] { 7.6, 3.0, 6.6, 2.1, 1, 0, 0 };
encodedData[106] = new double[] { 4.9, 2.5, 4.5, 1.7, 1, 0, 0 };
encodedData[107] = new double[] { 7.3, 2.9, 6.3, 1.8, 1, 0, 0 };
encodedData[108] = new double[] { 6.7, 2.5, 5.8, 1.8, 1, 0, 0 };
encodedData[109] = new double[] { 7.2, 3.6, 6.1, 2.5, 1, 0, 0 };
encodedData[110] = new double[] { 6.5, 3.2, 5.1, 2.0, 1, 0, 0 };
encodedData[111] = new double[] { 6.4, 2.7, 5.3, 1.9, 1, 0, 0 };
encodedData[112] = new double[] { 6.8, 3.0, 5.5, 2.1, 1, 0, 0 };
encodedData[113] = new double[] { 5.7, 2.5, 5.0, 2.0, 1, 0, 0 };
encodedData[114] = new double[] { 5.8, 2.8, 5.1, 2.4, 1, 0, 0 };
encodedData[115] = new double[] { 6.4, 3.2, 5.3, 2.3, 1, 0, 0 };
encodedData[116] = new double[] { 6.5, 3.0, 5.5, 1.8, 1, 0, 0 };
encodedData[117] = new double[] { 7.7, 3.8, 6.7, 2.2, 1, 0, 0 };
encodedData[118] = new double[] { 7.7, 2.6, 6.9, 2.3, 1, 0, 0 };
encodedData[119] = new double[] { 6.0, 2.2, 5.0, 1.5, 1, 0, 0 };
encodedData[120] = new double[] { 6.9, 3.2, 5.7, 2.3, 1, 0, 0 };
encodedData[121] = new double[] { 5.6, 2.8, 4.9, 2.0, 1, 0, 0 };
encodedData[122] = new double[] { 7.7, 2.8, 6.7, 2.0, 1, 0, 0 };
encodedData[123] = new double[] { 6.3, 2.7, 4.9, 1.8, 1, 0, 0 };
encodedData[124] = new double[] { 6.7, 3.3, 5.7, 2.1, 1, 0, 0 };
encodedData[125] = new double[] { 7.2, 3.2, 6.0, 1.8, 1, 0, 0 };
encodedData[126] = new double[] { 6.2, 2.8, 4.8, 1.8, 1, 0, 0 };
encodedData[127] = new double[] { 6.1, 3.0, 4.9, 1.8, 1, 0, 0 };
encodedData[128] = new double[] { 6.4, 2.8, 5.6, 2.1, 1, 0, 0 };
encodedData[129] = new double[] { 7.2, 3.0, 5.8, 1.6, 1, 0, 0 };
encodedData[130] = new double[] { 7.4, 2.8, 6.1, 1.9, 1, 0, 0 };
encodedData[131] = new double[] { 7.9, 3.8, 6.4, 2.0, 1, 0, 0 };
encodedData[132] = new double[] { 6.4, 2.8, 5.6, 2.2, 1, 0, 0 };
encodedData[133] = new double[] { 6.3, 2.8, 5.1, 1.5, 1, 0, 0 };
encodedData[134] = new double[] { 6.1, 2.6, 5.6, 1.4, 1, 0, 0 };
encodedData[135] = new double[] { 7.7, 3.0, 6.1, 2.3, 1, 0, 0 };
encodedData[136] = new double[] { 6.3, 3.4, 5.6, 2.4, 1, 0, 0 };
encodedData[137] = new double[] { 6.4, 3.1, 5.5, 1.8, 1, 0, 0 };
encodedData[138] = new double[] { 6.0, 3.0, 4.8, 1.8, 1, 0, 0 };
encodedData[139] = new double[] { 6.9, 3.1, 5.4, 2.1, 1, 0, 0 };
encodedData[140] = new double[] { 6.7, 3.1, 5.6, 2.4, 1, 0, 0 };
encodedData[141] = new double[] { 6.9, 3.1, 5.1, 2.3, 1, 0, 0 };
encodedData[142] = new double[] { 5.8, 2.7, 5.1, 1.9, 1, 0, 0 };
encodedData[143] = new double[] { 6.8, 3.2, 5.9, 2.3, 1, 0, 0 };
encodedData[144] = new double[] { 6.7, 3.3, 5.7, 2.5, 1, 0, 0 };
encodedData[145] = new double[] { 6.7, 3.0, 5.2, 2.3, 1, 0, 0 };
encodedData[146] = new double[] { 6.3, 2.5, 5.0, 1.9, 1, 0, 0 };
encodedData[147] = new double[] { 6.5, 3.0, 5.2, 2.0, 1, 0, 0 };
encodedData[148] = new double[] { 6.2, 3.4, 5.4, 2.3, 1, 0, 0 };
encodedData[149] = new double[] { 5.9, 3.0, 5.1, 1.8, 1, 0, 0 };
#endregion
var networkData = new NeuralNetworkData(rawData, ' ');
networkData.PrepareTrainingSet(
encodedData: encodedData,
trainDataPercentage: 0.80d,
seed: 72);
ANeuralNetwork nn = new ANeuralNetwork(
inputNodes: 4,
hiddenNodes: 7,
outputNodes: 3,
hiddenNodeActivation: Activation.HyperTan,
outputNodeActivation: Activation.Softmax);
nn.Train(
networkData.TrainingData,
maxEpochs: 1000,
learnRate: 0.05d,
momentum: 0.01d,
meanSquaredErrorThreshold: 0.040d);
Assert.IsTrue(nn.GetAccuracyFor(networkData.TrainingData) > 0.95d);
Assert.IsTrue(nn.GetAccuracyFor(networkData.TestData) > 0.95d);
Assert.AreEqual("virginica", networkData.DecodeOutput(nn.Predict(new double[] { 5.1, 3.5, 1.4, 0.2 })));
Assert.AreEqual("versicolor", networkData.DecodeOutput(nn.Predict(new double[] { 7.0, 3.2, 4.7, 1.4 })));
Assert.AreEqual("setosa", networkData.DecodeOutput(nn.Predict(new double[] { 7.4, 2.8, 6.1, 1.9 })));
}
