void AnalyzeMaxMin()
{
// Find max-min value for each column
analyzer = new MaxMinAnalyzer(Parameter.InputSize);
double[] input = new double[Parameter.InputSize];
DataSetReader reader = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
// Analyze training file.
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
{
analyzer.AnalyzeValue(i, input[i]);
}
}
// Analyze testing file.
reader = new DataSetReader(Parameter.TestingFile, Parameter.InputSize, Parameter.TargetSize);
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
{
analyzer.AnalyzeValue(i, input[i]);
}
}
}
public void Train()
{
// Read all training data and transforming them.
trainingData = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
double[][] data, target;
trainingData.ReadAll(out data, out target);
for (int i = 0; i < trainingData.Count; ++i)
{
// transform input level 1
for (int j = 0; j < Parameter.InputSize; ++j)
{
data[i][j] = inputTransformers1[j].TransformToValue(data[i][j]);
}
// transform input level 2
data[i] = inputTransformers2.TransformToDimension(data[i]);
// transform target
target[i] = targetTransform.TransformToDimension(target[i]);
}
oldArtAWeights = new List <double[]>();
bool isWeightsChanged = true, completedLimited = false;
// Order input and target pattern before.
if (param.Mode == ProcessModifiedHybridFuzzyARTMapMode.EXTERNAL ||
param.Mode == ProcessModifiedHybridFuzzyARTMapMode.DUAL)
{
mcoa.OrderSequence(data, target);
}
do
{
SaveOldWeights();
// Training ordered fuzzy art map
if (param.Mode == ProcessModifiedHybridFuzzyARTMapMode.EXTERNAL)
{
fartmap.Run(data, target);
}
else
{
mfartmap.Run(data, target);
}
double mseValue = CalculateMse();
if (mseValue < Parameter.LimitedMseValue)
{
isWeightsChanged = false;
}
performanceCounter.Increment();
if (param.LimitedEpochs != 0 &&
param.LimitedEpochs <= performanceCounter.Count)
{
completedLimited = true;
}
}while (!completedLimited && isWeightsChanged);
}
public void Test()
{
testData = new DataSetReader(Parameter.TestingFile, Parameter.InputSize,
Parameter.TargetSize);
outputData = new DataSetWriter(Parameter.OutputFile, false,
Parameter.InputSize, Parameter.TargetSize);
double[] data1 = new double[Parameter.InputSize]; // Original data from dataset file.
double[] data2 = new double[Parameter.InputSize]; // Temporaly transformed data (from data1).
double[] target = new double[Parameter.TargetSize];
double[] output = new double[1];
while (testData.Read(data1, target))
{
// transform input level 1
for (int i = 0; i < Parameter.InputSize; i++)
{
data2[i] = inputTransformers1[i].TransformToValue(data1[i]);
}
// transform input level 2
double[] dataInput = inputTransformers2.TransformToDimension(data2);
if (Parameter.Mode == ProcessModifiedHybridFuzzyARTMapMode.EXTERNAL)
{
fartmap.Run(dataInput);
output[0] = outputTransformers.TransformToValue(fartmap.Output);
}
else
{
mfartmap.Run(dataInput);
output[0] = outputTransformers.TransformToValue(mfartmap.Output);
}
outputData.Write(data1, output, target);
if (((int)output[0]) != 0)
{
if (((int)output[0]) == ((int)target[0]))
{
performanceCounter.RightIncrement();
}
else
{
performanceCounter.WrongIncrement();
}
}
else
{
performanceCounter.NonPredictIncrement();
}
}
outputData.Close();
}
public void TestWrite()
{
double[] dataBuffer = new double[] { 1.5 , 2.399 , 3.1234, 4.0555, 5.01111 };
double[] testBuffer = new double[] { 6.0 };
DataSetWriter dwr = new DataSetWriter(FILENAME, false, 5, 1);
for (int i = 0; i < 3; ++i)
{
dwr.Write(dataBuffer, testBuffer, testBuffer);
}
dwr.Close();
DataSetReader dsr = new DataSetReader(FILENAME, 5, 1);
for (;dsr.Read(dataBuffer, testBuffer);) ;
Assert.AreEqual(3, dsr.Count);
}
public void TestRead()
{
DataSetReader dsr = new DataSetReader(FILENAME, DATA_SIZE, TEST_SIZE);
int counter = 0;
List<double[]> result = new List<double[]>();
while (true)
{
// In testing, create new buffer for each iteration.
double[] dataBuffer = new double[DATA_SIZE];
double[] testBuffer = new double[TEST_SIZE];
if (!dsr.Read(dataBuffer, testBuffer))
break;
++counter;
}
Assert.AreEqual(3, counter);
}
public void Train()
{
// Read all training data and transforming them.
trainingData = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
double[][] data, target;
trainingData.ReadAll(out data, out target);
for (int i = 0; i < trainingData.Count; ++i)
{
// transform input level 1
for (int j = 0; j < Parameter.InputSize; ++j)
{
data[i][j] = inputTransformers1[j].TransformToValue(data[i][j]);
}
// transform input level 2
data[i] = inputTransformers2.TransformToDimension(data[i]);
// transform target
target[i] = targetTransform.TransformToDimension(target[i]);
}
oldArtAWeights = new List <double[]>();
bool isWeightsChanged = true, completedLimited = false;
do
{
SaveOldWeights();
// Training ordered fuzzy art map
network.Run(data, target);
double mseValue = CalculateMse();
if (mseValue < Parameter.LimitedMseValue)
{
isWeightsChanged = false;
}
performanceCounter.Increment();
if (parameter.LimitedEpochs != 0 &&
parameter.LimitedEpochs <= performanceCounter.Count)
{
completedLimited = true;
}
}while (!completedLimited && isWeightsChanged);
}
public void TestReadValidData()
{
DataSetReader dsr = new DataSetReader(FILENAME, DATA_SIZE, TEST_SIZE);
double[] dataBuffer = new double[DATA_SIZE];
double[] testBuffer = new double[TEST_SIZE];
bool success = dsr.Read(dataBuffer, testBuffer);
Assert.IsTrue(success);
double[] targetData = new double[] { 5.1000, 3.5000, 1.4000, 0.2000 };
double[] targetTest = new double[] { 1 };
Assert.AreEqual(targetData.Length, dataBuffer.Length);
Assert.AreEqual(targetTest.Length, testBuffer.Length);
for (int i = 0; i < targetData.Length; ++i)
Assert.AreEqual(targetData[i], dataBuffer[i], "Mismatch read data at " + (i + 1));
for (int i = 0; i < targetTest.Length; ++i)
Assert.AreEqual(targetTest[i], testBuffer[i], "Mismatch read test at " + (i + 1));
}
void AnalyzeMaxMin()
{
// Find max-min value for each column
analyzer = new MaxMinAnalyzer(Parameter.InputSize);
double[] input = new double[Parameter.InputSize];
DataSetReader reader = null;
if (File.Exists(Parameter.TrainingFile))
{
reader = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
}
else if (File.Exists(Parameter.TestingFile))
{
reader = new DataSetReader(Parameter.TestingFile,
parameter.InputSize, Parameter.TargetSize);
}
if (reader == null)
{
throw new ArgumentException("Process must have a trainging file or testing file.");
}
// Analyze training file.
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
{
analyzer.AnalyzeValue(i, input[i]);
}
}
// Analyze testing file.
reader = new DataSetReader(Parameter.TestingFile, Parameter.InputSize, Parameter.TargetSize);
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
{
analyzer.AnalyzeValue(i, input[i]);
}
}
}
public void Train()
{
oldArtAWeights = new List <double[]>();
bool isWeightsChanged = true, completedLimited = false;
do
{
SaveOldWeights();
trainingData = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
double[] data = new double[Parameter.InputSize];
double[] target = new double[Parameter.TargetSize];
while (trainingData.Read(data, target))
{
// transform input level 1
for (int i = 0; i < Parameter.InputSize; i++)
{
data[i] = inputTransformers1[i].TransformToValue(data[i]);
}
// transform input level 2
double[] dataInput = inputTransformers2.TransformToDimension(data);
// transform target
double[] targetInput = targetTransform.TransformToDimension(target);
network.Run(dataInput, targetInput);
}
double mseValue = CalculateMse();
if (mseValue < Parameter.LimitedMseValue)
{
isWeightsChanged = false;
}
performanceCounter.Increment();
if (parameter.LimitedEpochs != 0 &&
parameter.LimitedEpochs <= performanceCounter.Count)
{
completedLimited = true;
}
}while (!completedLimited && isWeightsChanged);
}
public void Test()
{
Performance.Reset();
testData = new DataSetReader(Parameter.TestingFile, Parameter.InputSize,
Parameter.TargetSize);
outputData = new DataSetWriter(Parameter.OutputFile, false,
Parameter.InputSize, Parameter.TargetSize);
double[] data1 = new double[Parameter.InputSize]; // Original data from dataset file.
double[] data2 = new double[Parameter.InputSize]; // Temporaly transformed data (from data1).
double[] target = new double[Parameter.TargetSize];
double[] output = new double[1];
while (testData.Read(data1, target))
{
// transform input level 1
for (int i = 0; i < Parameter.InputSize; i++)
{
data2[i] = inputTransformers1[i].TransformToValue(data1[i]);
}
// transform input level 2
double[] dataInput = inputTransformers2.TransformToDimension(data2);
network.Run(dataInput);
output[0] = outputTransformers.TransformToValue(network.Output);
outputData.Write(data1, output, target);
if (Math.Round(output[0], 1) == Math.Round(target[0], 1))
{
performanceCounter.RightIncrement();
}
else
{
performanceCounter.WrongIncrement();
}
}
outputData.Close();
}
public void Test()
{
testData = new DataSetReader(Parameter.TestingFile, Parameter.InputSize,
Parameter.TargetSize);
outputData = new DataSetWriter(Parameter.OutputFile, false,
Parameter.InputSize, Parameter.TargetSize);
double[] data1 = new double[Parameter.InputSize]; // Original data from dataset file.
double[] data2 = new double[Parameter.InputSize]; // Temporaly transformed data (from data1).
double[] target = new double[Parameter.TargetSize];
double[] output = new double[1];
while (testData.Read(data1, target))
{
// transform input level 1
for (int i = 0; i < Parameter.InputSize; i++)
{
data2[i] = inputTransformers1[i].TransformToValue(data1[i]);
}
// transform input level 2
double[] dataInput = inputTransformers2.TransformToDimension(data2);
if (Parameter.Mode == ProcessModifiedHybridFuzzyARTMapMode.EXTERNAL)
{
fartmap.Run(dataInput);
output[0] = outputTransformers.TransformToValue(fartmap.Output);
}
else
{
mfartmap.Run(dataInput);
output[0] = outputTransformers.TransformToValue(mfartmap.Output);
}
outputData.Write(data1, output, target);
if (((int)output[0]) != 0)
{
if (((int)output[0]) == ((int)target[0]))
performanceCounter.RightIncrement();
else
performanceCounter.WrongIncrement();
}
else
{
performanceCounter.NonPredictIncrement();
}
}
outputData.Close();
}
public void Train()
{
// Read all training data and transforming them.
trainingData = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
double[][] data, target;
trainingData.ReadAll(out data, out target);
for (int i = 0; i < trainingData.Count; ++i)
{
// transform input level 1
for (int j = 0; j < Parameter.InputSize; ++j)
{
data[i][j] = inputTransformers1[j].TransformToValue(data[i][j]);
}
// transform input level 2
data[i] = inputTransformers2.TransformToDimension(data[i]);
// transform target
target[i] = targetTransform.TransformToDimension(target[i]);
}
oldArtAWeights = new List<double[]>();
bool isWeightsChanged = true, completedLimited = false;
do
{
SaveOldWeights();
// Training ordered fuzzy art map
network.Run(data, target);
double mseValue = CalculateMse();
if (mseValue < Parameter.LimitedMseValue)
isWeightsChanged = false;
performanceCounter.Increment();
if (parameter.LimitedEpochs != 0
&& parameter.LimitedEpochs <= performanceCounter.Count)
{
completedLimited = true;
}
}
while (!completedLimited && isWeightsChanged);
}
public void Train()
{
// Read all training data and transforming them.
trainingData = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
double[][] data, target;
trainingData.ReadAll(out data, out target);
for (int i = 0; i < trainingData.Count; ++i)
{
// transform input level 1
for (int j = 0; j < Parameter.InputSize; ++j)
{
data[i][j] = inputTransformers1[j].TransformToValue(data[i][j]);
}
// transform input level 2
data[i] = inputTransformers2.TransformToDimension(data[i]);
// transform target
target[i] = targetTransform.TransformToDimension(target[i]);
}
oldArtAWeights = new List<double[]>();
bool isWeightsChanged = true, completedLimited = false;
// Order input and target pattern before.
if (param.Mode == ProcessModifiedHybridFuzzyARTMapMode.EXTERNAL ||
param.Mode == ProcessModifiedHybridFuzzyARTMapMode.DUAL)
{
mcoa.OrderSequence(data, target);
}
do
{
SaveOldWeights();
// Training ordered fuzzy art map
if (param.Mode == ProcessModifiedHybridFuzzyARTMapMode.EXTERNAL)
{
fartmap.Run(data, target);
}
else
{
mfartmap.Run(data, target);
}
double mseValue = CalculateMse();
if (mseValue < Parameter.LimitedMseValue)
{
isWeightsChanged = false;
}
performanceCounter.Increment();
if (param.LimitedEpochs != 0
&& param.LimitedEpochs <= performanceCounter.Count)
{
completedLimited = true;
}
}
while (!completedLimited && isWeightsChanged);
}
void AnalyzeMaxMin()
{
// Find max-min value for each column
analyzer = new MaxMinAnalyzer(Parameter.InputSize);
double[] input = new double[Parameter.InputSize];
DataSetReader reader = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
// Analyze training file.
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
analyzer.AnalyzeValue(i, input[i]);
}
// Analyze testing file.
reader = new DataSetReader(Parameter.TestingFile, Parameter.InputSize, Parameter.TargetSize);
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
analyzer.AnalyzeValue(i, input[i]);
}
}
void AnalyzeMaxMin()
{
// Find max-min value for each column
analyzer = new MaxMinAnalyzer(Parameter.InputSize);
double[] input = new double[Parameter.InputSize];
DataSetReader reader = null;
if (File.Exists(Parameter.TrainingFile))
{
reader = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
}
else if (File.Exists(Parameter.TestingFile))
{
reader = new DataSetReader(Parameter.TestingFile,
parameter.InputSize, Parameter.TargetSize);
}
if (reader == null)
throw new ArgumentException("Process must have a trainging file or testing file.");
// Analyze training file.
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
analyzer.AnalyzeValue(i, input[i]);
}
// Analyze testing file.
reader = new DataSetReader(Parameter.TestingFile, Parameter.InputSize, Parameter.TargetSize);
while (reader.Read(input, null))
{
for (int i = 0; i < Parameter.InputSize; i++)
analyzer.AnalyzeValue(i, input[i]);
}
}
public void Train()
{
oldArtAWeights = new List<double[]>();
bool isWeightsChanged = true, completedLimited = false;
do
{
SaveOldWeights();
trainingData = new DataSetReader(Parameter.TrainingFile,
Parameter.InputSize, Parameter.TargetSize);
double[] data = new double[Parameter.InputSize];
double[] target = new double[Parameter.TargetSize];
while (trainingData.Read(data, target))
{
// transform input level 1
for (int i = 0; i < Parameter.InputSize; i++)
{
data[i] = inputTransformers1[i].TransformToValue(data[i]);
}
// transform input level 2
double[] dataInput = inputTransformers2.TransformToDimension(data);
// transform target
double[] targetInput = targetTransform.TransformToDimension(target);
network.Run(dataInput, targetInput);
}
double mseValue = CalculateMse();
if (mseValue < Parameter.LimitedMseValue)
isWeightsChanged = false;
performanceCounter.Increment();
if (parameter.LimitedEpochs != 0
&& parameter.LimitedEpochs <= performanceCounter.Count)
completedLimited = true;
}
while (!completedLimited && isWeightsChanged);
}
public void Test()
{
Performance.Reset();
testData = new DataSetReader(Parameter.TestingFile, Parameter.InputSize,
Parameter.TargetSize);
outputData = new DataSetWriter(Parameter.OutputFile, false,
Parameter.InputSize, Parameter.TargetSize);
double[] data1 = new double[Parameter.InputSize]; // Original data from dataset file.
double[] data2 = new double[Parameter.InputSize]; // Temporaly transformed data (from data1).
double[] target = new double[Parameter.TargetSize];
double[] output = new double[1];
while (testData.Read(data1, target))
{
// transform input level 1
for (int i = 0; i < Parameter.InputSize; i++)
{
data2[i] = inputTransformers1[i].TransformToValue(data1[i]);
}
// transform input level 2
double[] dataInput = inputTransformers2.TransformToDimension(data2);
network.Run(dataInput);
output[0] = outputTransformers.TransformToValue(network.Output);
outputData.Write(data1, output, target);
if (Math.Round(output[0], 1) == Math.Round(target[0], 1))
performanceCounter.RightIncrement();
else
performanceCounter.WrongIncrement();
}
outputData.Close();
}