public void TrainHiddenMarkovModel()
{
//Trainigset
double[][] data = (from item in _trainingstore.Items select item.Data.ToArray()).ToArray();
var indexes = new int[data.Count()];
for (int i = 0; i < data.Count(); i++)
{
indexes[i] = i;
}
//Classifier
var density = new NormalDistribution(1);
_classifier = new ContinuousSequenceClassifier(_trainingstore.Items.Count, new Ergodic(ObservationLength),
density);
//Learn model
var teacher = new SequenceClassifierLearning(_classifier,
modelIndex =>
new ContinuousBaumWelchLearning(_classifier.Models[modelIndex])
{
Tolerance = _tolerance,
Iterations = _iterations
}
);
teacher.Run(data, indexes);
//Determine max likelihoods for all training items
double maxLikelihood;
foreach (TrainingItem item in _trainingstore.Items)
{
_classifier.Compute(item.Data.ToArray(), out maxLikelihood);
item.MaxLikelihood = maxLikelihood;
}
}
public void TrainHiddenMarkovModel()
{
//Trainigset
double[][] data = (from item in _trainingstore.Items select item.Data.ToArray()).ToArray();
var indexes = new int[data.Count()];
for (int i = 0; i < data.Count(); i++)
{
indexes[i] = i;
}
//Classifier
var density = new NormalDistribution(1);
_classifier = new ContinuousSequenceClassifier(_trainingstore.Items.Count, new Ergodic(ObservationLength),
density);
//Learn model
var teacher = new SequenceClassifierLearning(_classifier,
modelIndex =>
new ContinuousBaumWelchLearning(_classifier.Models[modelIndex])
{
Tolerance = _tolerance,
Iterations = _iterations
}
);
teacher.Run(data, indexes);
//Determine max likelihoods for all training items
double maxLikelihood;
foreach (TrainingItem item in _trainingstore.Items)
{
_classifier.Compute(item.Data.ToArray(), out maxLikelihood);
item.MaxLikelihood = maxLikelihood;
}
}
