class MAPDiagonalGMM

This class is a special case of a DiagonalGMM that implements the MAP algorithm instead of the EM algorithm.

Inheritance:

Public Fields

DiagonalGMM* prior_distribution

The prior distribution used in MAP

real weight_on_prior

The weight to give to the prior parameters during update

bool learn_weights

update Gaussian's weights

bool learn_variances

update Gaussian's variances

bool learn_means

update Gaussian's means

Public Methods

MAPDiagonalGMM(DiagonalGMM* prior_distribution_)

virtual void frameViterbiAccPosteriors(int t, real* inputs, real log_posterior)

The backward step of Viterbi for a frame

virtual void frameEMAccPosteriors(int t, real* inputs, real log_posterior)

The backward step of EM for a frame

virtual void eMUpdate()

The update after each iteration for EM

virtual void setDataSet(DataSet* data_)

Copy the parameters of the prior distribution

Inherited from DiagonalGMM:

Public Fields

int n_gaussians

real prior_weights

EMTrainer* initial_kmeans_trainer

MeasurerList* initial_kmeans_trainer_measurers

real* log_weights

real* dlog_weights

real* var_threshold

Sequence* log_probabilities_g

int best_gauss

Sequence* best_gauss_per_frame

real* sum_log_var_plus_n_obs_log_2_pi

real** minus_half_over_var

real** means_acc

Public Methods

virtual void reset()

virtual void setVarThreshold(real* var_threshold_)

virtual void eMIterInitialize()

virtual void iterInitialize()

virtual real frameLogProbability(int t, real* inputs)

virtual real viterbiFrameLogProbability(int t, real* inputs)

virtual real frameLogProbabilityOneGaussian(int g, real* inputs)

virtual void sequenceInitialize(Sequence* inputs)

virtual void eMSequenceInitialize(Sequence* inputs)

virtual void update()

virtual void frameBackward(int t, real* f_inputs, real* beta_, real* f_outputs, real* alpha_)

virtual void frameDecision(int t, real* decision)

Inherited from Distribution:

Public Fields

real log_probability

Sequence* log_probabilities

Public Methods

virtual real logProbability(Sequence* inputs)

virtual real viterbiLogProbability(Sequence* inputs)

virtual void eMAccPosteriors(Sequence* inputs, real log_posterior)

virtual void viterbiAccPosteriors(Sequence* inputs, real log_posterior)

virtual void decode(Sequence* inputs)

virtual void eMForward(Sequence* inputs)

virtual void viterbiForward(Sequence* inputs)

virtual void viterbiBackward(Sequence* inputs, Sequence* alpha)

Public Members

Returns the decision of the distribution

Inherited from GradientMachine:

Public Fields

int n_inputs

int n_outputs

Parameters* params

Parameters* der_params

Sequence* beta

Public Methods

virtual void forward(Sequence* inputs)

virtual void backward(Sequence* inputs, Sequence* alpha)

virtual void setPartialBackprop(bool flag=true)

virtual void frameForward(int t, real* f_inputs, real* f_outputs)

virtual void loadXFile(XFile* file)

virtual void saveXFile(XFile* file)

Inherited from Machine:

Public Fields

Sequence* outputs

Inherited from Object:

Public Fields

Allocator* allocator

Public Methods

void addOption(const char* name, int size, void* ptr, const char* help="")

void addIOption(const char* name, int* ptr, int init_value, const char* help="")

void addROption(const char* name, real* ptr, real init_value, const char* help="")

void addBOption(const char* name, bool* ptr, bool init_value, const char* help="")

void addOOption(const char* name, Object** ptr, Object* init_value, const char* help="")

void setOption(const char* name, void* ptr)

void setIOption(const char* name, int option)

void setROption(const char* name, real option)

void setBOption(const char* name, bool option)

void setOOption(const char* name, Object* option)

void load(const char* filename)

void save(const char* filename)

void* operator new(size_t size, Allocator* allocator_=NULL)

void* operator new(size_t size, Allocator* allocator_, void* ptr_)

void operator delete(void* ptr)

Documentation

This class is a special case of a DiagonalGMM that implements the MAP algorithm instead of the EM algorithm. This means that the mean parameters will be changed according to the Maximum A Posteriori algorithm, given a prior value of the means (through a prior DiagonalGMM given in the constructor). Moreover, the variances and weights are not changed, as experimental results tend to show that there is no effects when they are changed.

DiagonalGMM* prior_distribution

The prior distribution used in MAP

real weight_on_prior

The weight to give to the prior parameters during update

bool learn_weights

update Gaussian's weights

bool learn_variances

update Gaussian's variances

bool learn_means

update Gaussian's means

MAPDiagonalGMM(DiagonalGMM* prior_distribution_)

virtual void frameViterbiAccPosteriors(int t, real* inputs, real log_posterior)

The backward step of Viterbi for a frame

virtual void frameEMAccPosteriors(int t, real* inputs, real log_posterior)

The backward step of EM for a frame

virtual void eMUpdate()

The update after each iteration for EM

virtual void setDataSet(DataSet* data_)

Copy the parameters of the prior distribution

This class has no child classes.

Author:

Samy Bengio (bengio@idiap.ch) Johnny Mariethoz (Johnny.Mariethoz@idiap.ch)

Alphabetic index HTML hierarchy of classes or Java