public ITransformer TrainModel(TrainingAlgorithm algorithm, DataControl.TrainingOptions options)
{
switch (algorithm)
{
case TrainingAlgorithm.LOGISTIC_REGRESSION:
return(LogisticRegression(options));
case TrainingAlgorithm.NAIVE_BAYES:
return(NaiveBayes(options));
case TrainingAlgorithm.BINARY_STOCHASTIC_DUAL_COORDINATE_ASCENT:
return(BinaryStochasticDualCoordinateAscent(options));
case TrainingAlgorithm.FAST_TREE:
return(FastTree(options));
case TrainingAlgorithm.STOCHASTIC_DUAL_COORDINATE_ASCENT:
return(StochasticDualCoordinateAscent(options));
case TrainingAlgorithm.STOCHASTIC_GRADIENT_DESCENT:
return(StochasticGradientDescent(options));
default:
return(null);
}
}
public void Evaluate(ITransformer model, TrainingAlgorithm algorithm, string labelColumn)
{
switch (algorithm)
{
case TrainingAlgorithm.STOCHASTIC_DUAL_COORDINATE_ASCENT:
case TrainingAlgorithm.LOGISTIC_REGRESSION:
case TrainingAlgorithm.NAIVE_BAYES:
EvaluateMulticlass(model, labelColumn);
break;
case TrainingAlgorithm.BINARY_STOCHASTIC_DUAL_COORDINATE_ASCENT:
case TrainingAlgorithm.FAST_TREE:
case TrainingAlgorithm.STOCHASTIC_GRADIENT_DESCENT:
EvaluateBinary(model, labelColumn);
break;
}
}
static void Main(string[] args)
{
// words to be tested
List<int> wordsToBeTested = new List<int>();
for(int i = 1; i <= 100; i++)
wordsToBeTested.Add(i);
//wordsToBeTested.Add(1);
// samples to be tested
List<int> samplesToBeTested = new List<int>();
samplesToBeTested.Add(1);
samplesToBeTested.Add(2);
samplesToBeTested.Add(3);
samplesToBeTested.Add(4);
samplesToBeTested.Add(5);
// audio directory
List<string> audioDirectory = new List<string>();
audioDirectory.Add("C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\audio\\092910_123758_Hebrew");
audioDirectory.Add("C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\audio\\101410_140344_Hebrew");
audioDirectory.Add("C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\audio\\101510_111237_Hebrew");
// audio file names
List<string> audioFileNames = new List<string>();
audioFileNames.Add("[Wed_(Sep_29_2010)_12-37-58]_4124143701_");
audioFileNames.Add("[Thu_(Oct_14_2010)_14-03-44]_4122688595_");
audioFileNames.Add("[Fri_(Oct_15_2010)_11-12-37]_4126203298_");
for (int i = 0; i < 3; i++)
{
List<int> speakersToBeTested = new List<int>();
speakersToBeTested.Add(i);
// set up data
data = new Data(audioDirectory, audioFileNames, 5, speakersToBeTested, wordsToBeTested, samplesToBeTested,
"C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\config_files\\config.txt.100.english");
// setup grammar
GrammarCreator gc = new GrammarCreator("C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data", "Hebrew", "allcombinations");
// setup training
TrainingAlgorithm ta = new TrainingAlgorithm(gc, data, 10, 15);
ta.LearnAllWords();
}
}
static void Main(string[] args)
{
// speakers to be tested
List<int> speakersToBeTested = new List<int>();
speakersToBeTested.Add(0);
speakersToBeTested.Add(1);
speakersToBeTested.Add(2);
// words to be tested
List<int> wordsToBeTested = new List<int>();
for (int i = 1; i <= 50; i += 5)
wordsToBeTested.Add(i);
//wordsToBeTested.Add(25);
// samples to be tested
List<int> samplesToBeTested = new List<int>();
samplesToBeTested.Add(1);
samplesToBeTested.Add(2);
samplesToBeTested.Add(3);
samplesToBeTested.Add(4);
samplesToBeTested.Add(5);
// audio directory
List<string> audioDirectory = new List<string>();
audioDirectory.Add("C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\audio\\092910_123758_Hebrew");
audioDirectory.Add("C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\audio\\101410_140344_Hebrew");
audioDirectory.Add("C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\audio\\101510_111237_Hebrew");
// audio file names
List<string> audioFileNames = new List<string>();
audioFileNames.Add("[Wed_(Sep_29_2010)_12-37-58]_4124143701_");
audioFileNames.Add("[Thu_(Oct_14_2010)_14-03-44]_4122688595_");
audioFileNames.Add("[Fri_(Oct_15_2010)_11-12-37]_4126203298_");
//GrammarReader gr = new GrammarReader("Hebrew", "C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data", "Allphone");
// set up data
data = new Data(audioDirectory, audioFileNames, 5, speakersToBeTested, wordsToBeTested, samplesToBeTested,
"C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data\\config_files\\config.txt.100.english");
TrainingAlgorithm ta = new TrainingAlgorithm(10, data, directory, MWordTypes, MSamples, MSpeakers, "C:\\Users\\Administrator\\Documents\\Visual Studio 2008\\Projects\\cmuspeechrecognition_cmuspeechmain\\test_data", 1, "eliminative");
ta.trainWordsDiscriminativeEliminative();
}
/// <summary>
/// Given a list of speakers and the samples per speakers to train on, training is carried out to produce a List of pronunciations
/// </summary>
/// <param name="speakerList">list of speakers for training</param>
/// <param name="sampleList">list of samples for training</param>
/// <returns>List of pronunciations</returns>
private Dictionary<string, List<string>> trainAllWords(List<int> speakerList, List<int> sampleList)
{
string speaker = "speaker-";
for (int i = 0; i < speakerList.Count; i++)
{
speaker += speakerList[i].ToString()+"_";
}
string sample = "sample-";
List<int> trueSampleList = new List<int>();
trueSampleList.AddRange(sampleList);
for (int i = 0; i < trueSampleList.Count; i++)
{
trueSampleList[i]++;
sample += trueSampleList[i].ToString() + "_";
}
System.Diagnostics.Debug.WriteLine("training " + speaker + sample);
// words to be tested
List<int> wordsToBeTested = new List<int>();
for (int i = 1; i <= MWordTypes; i++)
wordsToBeTested.Add(i);
// set up data
Data tempData = new Data(data.audioDirectory, data.audioFileName, data.numberOfSamplesPerWord, speakerList, wordsToBeTested, trueSampleList,
data.wordListPath);
// setup grammar
GrammarCreator gc = new GrammarCreator(grammarDirectory, data.language + speaker + sample, "allcombinations");
// setup training
TrainingAlgorithm ta = new TrainingAlgorithm(gc, tempData, numberOfAlternates, 15);
return ta.LearnAllWords();
}
/// <summary>
/// Training callback, invoked at each iteration
/// </summary>
/// <param name="epoch">Epoch number</param>
/// <param name="error">Current error</param>
/// <param name="algorithm">Training algorithm</param>
private void TrainingCallback(int epoch, double error, TrainingAlgorithm algorithm)
{
Invoke(addAction, new object [] { epoch, error, algorithm, _dgvTrainingResults });
}
public void set_training_algorithm(TrainingAlgorithm training_algorithm)
{
fannfixedPINVOKE.neural_net_set_training_algorithm(swigCPtr, (int)training_algorithm);
}
public TrainingAlgorithm get_training_algorithm()
{
TrainingAlgorithm ret = (TrainingAlgorithm)fannfixedPINVOKE.neural_net_get_training_algorithm(swigCPtr);
return(ret);
}
void Start()
{
ga = GetComponent<TrainingAlgorithm>();
spawnWave(1);
timer = 30.0f;
}
static void Main()
{
const float desired_error = 0.0F;
uint max_neurons = 30;
uint neurons_between_reports = 1;
uint bit_fail_train, bit_fail_test;
float mse_train, mse_test;
DataType[] output;
DataType[] steepness = new DataType[1];
int multi = 0;
ActivationFunction[] activation = new ActivationFunction[1];
TrainingAlgorithm training_algorithm = TrainingAlgorithm.TRAIN_RPROP;
Console.WriteLine("Reading data.");
using (TrainingData trainData = new TrainingData("..\\..\\..\\datasets\\parity8.train"))
using (TrainingData testData = new TrainingData("..\\..\\..\\datasets\\parity8.test"))
{
trainData.ScaleTrainData(-1, 1);
testData.ScaleTrainData(-1, 1);
Console.WriteLine("Creating network.");
using (NeuralNet net = new NeuralNet(NetworkType.SHORTCUT, 2, trainData.InputCount, trainData.OutputCount))
{
net.TrainingAlgorithm = training_algorithm;
net.ActivationFunctionHidden = ActivationFunction.SIGMOID_SYMMETRIC;
net.ActivationFunctionOutput = ActivationFunction.LINEAR;
net.TrainErrorFunction = ErrorFunction.ERRORFUNC_LINEAR;
if (multi == 0)
{
steepness[0] = 1;
net.CascadeActivationSteepnesses = steepness;
activation[0] = ActivationFunction.SIGMOID_SYMMETRIC;
net.CascadeActivationFunctions = activation;
net.CascadeCandidateGroupsCount = 8;
}
if (training_algorithm == TrainingAlgorithm.TRAIN_QUICKPROP)
{
net.LearningRate = 0.35F;
net.RandomizeWeights(-2.0F, 2.0F);
}
net.BitFailLimit = (DataType)0.9;
net.TrainStopFunction = StopFunction.STOPFUNC_BIT;
net.PrintParameters();
net.Save("..\\..\\..\\examples\\cascade_train2.net");
Console.WriteLine("Training network.");
net.CascadetrainOnData(trainData, max_neurons, neurons_between_reports, desired_error);
net.PrintConnections();
mse_train = net.TestData(trainData);
bit_fail_train = net.BitFail;
mse_test = net.TestData(testData);
bit_fail_test = net.BitFail;
Console.WriteLine("\nTrain error: {0}, Train bit-fail: {1}, Test error: {2}, Test bit-fail: {3}\n",
mse_train, bit_fail_train, mse_test, bit_fail_test);
for (int i = 0; i < trainData.TrainDataLength; i++)
{
output = net.Run(trainData.GetTrainInput((uint)i));
if ((trainData.GetTrainOutput((uint)i)[0] >= 0 && output[0] <= 0) ||
(trainData.GetTrainOutput((uint)i)[0] <= 0 && output[0] >= 0))
{
Console.WriteLine("ERROR: {0} does not match {1}", trainData.GetTrainOutput((uint)i)[0], output[0]);
}
}
Console.WriteLine("Saving network.");
net.Save("..\\..\\..\\examples\\cascade_train.net");
Console.ReadKey();
}
}
}
public void set_training_algorithm(TrainingAlgorithm training_algorithm)
{
fanndoublePINVOKE.neural_net_set_training_algorithm(swigCPtr, (int)training_algorithm);
}
