backward (
        self,
        emissionSequence,
        scalingVector,
        )

Result: the (N x T)-matrix containing the backward-variables

baumWelch (
        self,
        trainingSequences,
        nrSteps,
        loglikelihoodCutoff,
        )

Reestimate the model parameters given the training_sequences. Perform at most nr_steps until the improvement in likelihood is below likelihood_cutoff

training_sequences can either be a SequenceSet or a Sequence

Result: Final loglikelihood

baumWelchDelete ( self )

Delete the necessary temporary variables for Baum-Welch-reestimation

baumWelchSetup (
        self,
        trainingSequences,
        nrSteps,
        )

Setup necessary temporary variables for Baum-Welch-reestimation. Use baum_welch_setup and baum_welch_step if you want more control over the training, compute diagnostics or do noise-insertion

training_sequences can either be a SequenceSet or a Sequence

Return: a C-array of type smosqd_t

baumWelchStep (
        self,
        nrSteps,
        loglikelihoodCutoff,
        )

Compute one iteration of Baum Welch estimation. Use baum_welch_setup and baum_welch_step if you want more control over the training, compute diagnostics or do noise-insertion

training_sequences can either be a SequenceSet or a Sequence

forward ( self,  emissionSequence )

Result: the (N x T)-matrix containing the forward-variables and the scaling vector

getEmission ( self,  i )

Return (mu, sigma^2)

getTransition (
        self,
        i,
        j,
        )

Returns the probability of the transition from state i to state j.

Raises IndexError if the transition is not allowed

loglikelihoods ( self,  emissionSequences )

Compute a vector ( log( P[s| model]) )_{s} of log-likelihoods of the individual emission_sequences using the forward algorithm

emission_sequences is of type SequenceSet

Result: log( P[emissionSequences| model]) of type float (numarray) vector of floats

normalize ( self )

Normalize transition probs, emission probs (if applicable)

pathPosterior (
        self,
        sequence,
        path,
        )

Returns the log posterior probability for path having generated sequence.

CAVEAT: statePosterior needs to calculate the complete forward and backward matrices. If you are interested in multiple paths it would be more efficient to use the posterior function directly and not multiple calls to pathPosterior

posterior ( self,  sequence )

Posterior distribution matrix for sequence.

sample (
        self,
        seqNr,
        T,
        seed=-1,
        )

Sample emission sequences

different function signatures require overloading of parent class methods

sampleSingle (
        self,
        T,
        seed=-1,
        )

Sample a single emission sequence of length at most T. Returns a Sequence object.

setEmission (
        self,
        i,
        (,
        )

Set the emission distributionParameters for state i

setTransition (
        self,
        i,
        j,
        prob,
        )

Accessor function for the transition a_ij

statePosterior (
        self,
        sequence,
        state,
        time,
        )

Return the log posterior probability for being at state at time time in sequence.

CAVEAT: statePosterior needs to calculate the complete forward and backward matrices. If you are interested in multiple states it would be more efficient to use the posterior function directly and not multiple calls to statePosterior

toMatrices ( self )

Return the parameters in matrix form.

viterbi ( self,  emissionSequences )

Compute the Viterbi-path for each sequence in emissionSequences

emission_sequences can either be a SequenceSet or an EmissionSequence

Result: [q_0, ..., q_T] the viterbi-path of emission_sequences is an emmissionSequence object, [[q_0^0, ..., q_T^0], ..., [q_0^k, ..., q_T^k]} for a k-sequence SequenceSet