34#ifndef STK_IMIXTURELEARNER_H
35#define STK_IMIXTURELEARNER_H
92 template<
class ColVector>
98 template<
class ColVector,
class RowVector>
112template<
class ColVector>
117 for (
int i=
tik_.beginRows();
i <
tik_.endRows();
i++)
128template<
class ColVector,
class RowVector>
134 for (
int i=
tik_.beginRows();
i <
tik_.endRows();
i++)
In this file we define the abstract base class for mixture statistical models.
This file contain the functors computings statistics.
Base class for Learner of a Mixture mixed model.
CVectorXi ziPred_
The predicted class label.
virtual void paramUpdateStep()=0
Compute the model parameters given the current mixture parameters and imputation/simulation of the mi...
void setState(Clust::modelState state)
set the state of the model : should be used by any strategy
virtual void mapStep()
Compute ziPred using the Map estimate.
void setClassLabels(ColVector const &zi)
set the mixture parameters using the given class labels.
Clust::modelState state() const
Clust::modelState state_
state of the model
CVectorXi const & ziPred() const
virtual ~IMixtureLearner()
destructor
Interface base class for Mixture (composed) model.
CArrayXX tik_
The tik probabilities.
CVectorXi const & zi() const
CPointX const & pk() const
CVectorXi zi_
The zi class label.
CPointX pk_
The proportions of each mixtures.
CPointX tk_
The sum of the columns of tik_.
The MultidimRegression class allows to regress a multidimensional output variable among a multivariat...
modelState
Give the state of the model.
hidden::FunctorTraits< Derived, SumOp >::Row sumByCol(Derived const &A)
The namespace STK is the main domain space of the Statistical ToolKit project.