44 , nbIterBurn_(0), nbIterLong_(0), epsilon_(0.) {}
49 , p_model_(
algo.p_model_)
50 , nbIterBurn_(
algo.nbIterBurn_)
51 , nbIterLong_(
algo.nbIterLong_)
52 , epsilon_(
algo.epsilon_)
68#ifdef STK_MIXTURE_VERY_VERBOSE
69 stk_cout <<
_T(
"-------------------------------\n");
70 stk_cout <<
_T(
"Entering IMixtureAlgoPredict::predictBayesClassifier()\n");
84#ifdef STK_MIXTURE_VERBOSE
92#ifdef STK_MIXTURE_VERBOSE
103#ifdef STK_MIXTURE_VERY_VERBOSE
104 stk_cout <<
_T(
"-------------------------------\n");
105 stk_cout <<
_T(
"Entering IMixtureAlgoPredict::burnStep()\n");
121#ifdef STK_MIXTURE_VERBOSE
In this file we define the interface base class for mixture predicting algorithms.
In this file we define the abstract base class for mixture models.
#define stk_cout
Standard stk output stream.
#define _T(x)
Let x unmodified.
Sdk class for all library Exceptions.
Interface base class for predicting algorithms.
bool predictBayesClassifier()
predict class labels when there is no missing values.
IMixtureAlgoPredict()
default constructor
virtual ~IMixtureAlgoPredict()
destructor
int nbIterLong_
maximal number of iterations of the algorithm
bool burnStep()
Perform burn step using SEM algorithm.
void setModel(IMixtureComposer *p_model)
set model
IMixtureComposer * p_model_
pointer on the mixture model
Base class for Mixture (composed) model.
virtual void initializeStep()
Initialize the model before its first use.
virtual Real eStep()
compute the zi, the lnLikelihood of the current estimates and the next value of the tik.
virtual void mapStep()
Compute zi using the Map estimate.
virtual int sStep()
Simulate zi accordingly to tik and replace tik by zik by calling cStep().
virtual void finalizeStep()
Finalize the estimation of the model.
virtual void setParametersStep()
Utility method allowing to signal to a mixture to set its parameters.
virtual void storeIntermediateResults(int iteration)
This step can be used to signal to the mixtures that they must store results.
virtual void samplingStep()
Simulation of all the latent variables and/or missing data excluding class labels.
Abstract base class for all classes having a.
String msg_error_
String with the last error message.
String const & error() const
get the last error message.
The MultidimRegression class allows to regress a multidimensional output variable among a multivariat...
String exceptionToString(exceptions const &type)
convert a Clust::exceptions to a String.
exceptions
Specific exceptions allowing to handle the erroros that can occur in the estimation process.
The namespace STK is the main domain space of the Statistical ToolKit project.