public void Predict_VideoViewsSeriesAndRegularErgodicModelAnd30DaysTestSequence_PredictionResult()
{
var util = new TestPipedDataUtils();
var series = util.GetSvcData(util.VideoViewsFilePath, new DateTime(2014, 01, 01), new DateTime(2014, 05, 31));
var test = util.GetSvcData(util.VideoViewsFilePath, new DateTime(2014, 06, 01), new DateTime(2014, 06, 30));
var model = (HiddenMarkovModelMixtureDistribution)HiddenMarkovModelFactory.GetModel <HiddenMarkovModelMixtureDistribution, Mixture <IMultivariateDistribution> >(
new ModelCreationParameters <Mixture <IMultivariateDistribution> >()
{
NumberOfComponents = _NumberOfComponents,
NumberOfStates = _NumberOfStates
});
model.Normalized = true;
model.Train(series, _NumberOfIterations, _LikelihoodTolerance);
var result = model.Predict(PredictorType.HmmLikelihood, series, test, 30, _NumberOfIterations, _LikelihoodTolerance);
util.SaveToFile("regular_data.csv", test, result.Predicted);
Assert.IsNotNull(result);
Assert.AreEqual(30, result.Predicted.Length);
}
public void Predict_ModelAndPredictionRequest_PredictedValue()
{
var util = new TestPipedDataUtils();
var series = util.GetSvcData(util.VideoViewsFilePath, new DateTime(2014, 01, 01), new DateTime(2014, 06, 01));
var model = (HiddenMarkovModelMixtureDistribution)HiddenMarkovModelFactory.GetModel(new ModelCreationParameters <Mixture <IMultivariateDistribution> >()
{
NumberOfComponents = _NumberOfComponents, NumberOfStates = _NumberOfStates
});
model.Normalized = true;
model.Train(series, _NumberOfIterations, _LikelihoodTolerance);
var request = new SearchBasedPredictionRequest();
request.TrainingSet = series;
request.NumberOfDays = 10;
var result = model.Predict(PredictorType.HmmLikelihood, series, _NumberOfIterations, _LikelihoodTolerance);
Assert.AreEqual(0, 0);//result.Predicted);
}
