class Distribution

This class is designed to handle generative distribution models such as Gaussian Mixture Models and Hidden Markov Models.

Inheritance:

Public Fields

real log_probability

the log likelihood

Sequence* log_probabilities

the log likelihood for each frame when available

Public Methods

Distribution(int n_inputs_, int n_params_=0)

virtual real logProbability(Sequence* inputs)

Returns the log probability of a sequence represented by inputs

virtual real viterbiLogProbability(Sequence* inputs)

Returns the viterbi score of a sequence represented by inputs

virtual real frameLogProbability(int t, real* f_inputs)

Returns the log probability of a frame of a sequence

virtual real viterbiFrameLogProbability(int t, real* f_inputs)

Returns the log probability of a frame of a sequence on viterbi mode

virtual void eMIterInitialize()

Methods used to initialize the model at the beginning of each EM iteration

virtual void iterInitialize()

Methods used to initialize the model at the beginning of each gradient descent iteration

virtual void eMSequenceInitialize(Sequence* inputs)

Methods used to initialize the model at the beginning of each example during EM training

virtual void sequenceInitialize(Sequence* inputs)

Methods used to initialize the model at the beginning of each example during gradient descent training

virtual void eMAccPosteriors(Sequence* inputs, real log_posterior)

The backward step of EM for a sequence

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

The backward step of EM for a frame

virtual void viterbiAccPosteriors(Sequence* inputs, real log_posterior)

The backward step of Viterbi learning for a sequence

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

The backward step of Viterbi for a frame

virtual void eMUpdate()

The update after each iteration for EM

virtual void update()

The update after each gradient iteration

virtual void decode(Sequence* inputs)

For some distribution like SpeechHMM, decodes the most likely path

virtual void eMForward(Sequence* inputs)

Same as forward, but for EM

virtual void viterbiForward(Sequence* inputs)

Same as forward, but for Viterbi

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

Same as backward, but for one frame only

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

Same as backward, but for Viterbi

virtual void frameDecision(int t, real* decision)

Returns the decision of a frame of a sequence

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

Public Methods

virtual void reset()

virtual void setDataSet(DataSet* dataset_)

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 designed to handle generative distribution models such as Gaussian Mixture Models and Hidden Markov Models. As distribution inherits from GradientMachine, they can be trained by gradient descent or by Expectation Maximization (EM) or even Viterbi.

Note that the output of a distribution is the negative log likelihood.

real log_probability

the log likelihood

Sequence* log_probabilities

the log likelihood for each frame when available

Distribution(int n_inputs_, int n_params_=0)

virtual real logProbability(Sequence* inputs)

Returns the log probability of a sequence represented by inputs

virtual real viterbiLogProbability(Sequence* inputs)

Returns the viterbi score of a sequence represented by inputs

virtual real frameLogProbability(int t, real* f_inputs)

Returns the log probability of a frame of a sequence

virtual real viterbiFrameLogProbability(int t, real* f_inputs)

Returns the log probability of a frame of a sequence on viterbi mode

virtual void eMIterInitialize()

Methods used to initialize the model at the beginning of each EM iteration

virtual void iterInitialize()

Methods used to initialize the model at the beginning of each gradient descent iteration

virtual void eMSequenceInitialize(Sequence* inputs)

Methods used to initialize the model at the beginning of each example during EM training

virtual void sequenceInitialize(Sequence* inputs)

Methods used to initialize the model at the beginning of each example during gradient descent training

virtual void eMAccPosteriors(Sequence* inputs, real log_posterior)

The backward step of EM for a sequence

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

The backward step of EM for a frame

virtual void viterbiAccPosteriors(Sequence* inputs, real log_posterior)

The backward step of Viterbi learning for a sequence

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

The backward step of Viterbi for a frame

virtual void eMUpdate()

The update after each iteration for EM

virtual void update()

The update after each gradient iteration

virtual void decode(Sequence* inputs)

For some distribution like SpeechHMM, decodes the most likely path

virtual void eMForward(Sequence* inputs)

Same as forward, but for EM

virtual void viterbiForward(Sequence* inputs)

Same as forward, but for Viterbi

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

Same as backward, but for one frame only

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

Same as backward, but for Viterbi

Returns the decision of the distribution

Returns the decision of the distribution

virtual void frameDecision(int t, real* decision)

Returns the decision of a frame of a sequence

Direct child classes:

TableLookupDistribution
SpeechMLPDistr
ParzenDistribution
Multinomial
HMM
DiagonalGMM

Author:

Samy Bengio (bengio@idiap.ch)

Alphabetic index HTML hierarchy of classes or Java