Version 4.0.0
Main Page | Class Hierarchy | Class List | File List | Class Members | Related Pages

MTDMarkov Class Reference

Mixture Transition Distribution Markov modelling, estimation and simulation. More...

#include <seqpp/MTDMarkov.h>

Inheritance diagram for MTDMarkov:

Markov PhasedMarkov List of all members.

Public Member Functions

template<class TSeq>
 MTDMarkov (const TSeq &tseq, short mkv_order, short nbseed=NBSEED, int nbiter_max=NBITERMAX, double eps=EPS, bool log=false)
 Constructor 1 from a SequenceSet or a Sequence.
 MTDMarkov (unsigned long **count, short size, short mtd_order, short mkv_order, short nbseed=NBSEED, int nbiter_max=NBITERMAX, double eps=EPS, bool log=false)
 Constructor 2 from a coded-word count.
void estimate (unsigned long **count, bool decal_required, short mkv_order, short nbseed, int nbiter_max, double eps, bool log)
 performs the estimation [used by each constructor]
 ~MTDMarkov ()
 Destructor.

Protected Attributes

mtd_core * _mtdcore
 Parcimonious Context Trees.

Detailed Description

Mixture Transition Distribution Markov modelling, estimation and simulation.

MTDMarkov is a Markov object with a different estimation step. This object performs the estimation with the Mixture Transition Distribution algorithm and then transfoms, once per phase, the MTD model in a markovian matrix.


Constructor & Destructor Documentation

template<class TSeq>
MTDMarkov::MTDMarkov const TSeq &  tseq,
short  mkv_order,
short  nbseed = NBSEED,
int  nbiter_max = NBITERMAX,
double  eps = EPS,
bool  log = false
[inline]
 

Constructor 1 from a SequenceSet or a Sequence.

Parameters:
tseq a set of sequences or a sequence for estimation
mkv_order markovian order on the markovian matrix of the MTD
nbseed number of seeds for the EM algorithm
nbiter_max maximum iterations number of the EM algorithm
eps value of the epsilon of the EM algorithm
log true to log the successive likelihood values

MTDMarkov::MTDMarkov unsigned long **  count,
short  size,
short  mtd_order,
short  mkv_order,
short  nbseed = NBSEED,
int  nbiter_max = NBITERMAX,
double  eps = EPS,
bool  log = false
[inline]
 

Constructor 2 from a coded-word count.

Parameters:
count count of all the coded word(base size) of size order+1 for each phase, for estimation
size alphabet size
mtd_order markovian order for the resulting model
mkv_order markovian order on the markovian matrices used in the MTD
nbseed number of seeds for the EM algorithm
nbiter_max maximum iterations number of the EM algorithm
eps value of the epsilon of the EM algorithm
log true to log the successive likelihood values


The documentation for this class was generated from the following file:



Download seq++ 4.0.0
Download previous versions
Statistique & Genome Home


Generated on Sun Apr 3 14:16:10 2005 for seqpp by doxygen 1.3.9.1