Package | Description |
---|---|
de.jstacs.sequenceScores.statisticalModels.trainable.hmm |
The package provides all interfaces and classes for a hidden Markov model (HMM).
|
de.jstacs.sequenceScores.statisticalModels.trainable.hmm.models |
The package provides different implementations of hidden Markov models based on
AbstractHMM . |
Modifier and Type | Method and Description |
---|---|
AbstractHMM |
AbstractHMM.clone() |
static AbstractHMM |
HMMFactory.createErgodicHMM(HMMTrainingParameterSet pars,
int order,
double ess,
double selfTranistionPart,
double expectedSequenceLength,
Emission... emission)
This method creates an ergodic, i.e.
|
static AbstractHMM |
HMMFactory.createProfileHMM(MaxHMMTrainingParameterSet trainingParameterSet,
double[][] initFromTo,
boolean likelihood,
int order,
int numLayers,
AlphabetContainer con,
double ess,
boolean conditionalMain,
boolean closeCircle,
double[][] conditionInitProbs,
boolean insertUniform)
Creates a new profile HMM for a given architecture and number of layers.
|
static AbstractHMM |
HMMFactory.createProfileHMM(MaxHMMTrainingParameterSet trainingParameterSet,
double[][] initFromTo,
boolean likelihood,
int order,
int numLayers,
AlphabetContainer con,
double ess,
boolean conditionalMain,
int joiningStates,
double[][] conditionInitProbs,
boolean insertUniform)
Creates a new profile HMM for a given architecture and number of layers.
|
static AbstractHMM |
HMMFactory.createProfileHMM(MaxHMMTrainingParameterSet trainingParameterSet,
HMMFactory.HMMType type,
boolean likelihood,
int order,
int numLayers,
AlphabetContainer con,
double ess,
boolean conditionalMain,
boolean closeCircle,
double[][] conditionInitProbs)
Creates a new profile HMM for a given architecture and number of layers.
|
static AbstractHMM |
HMMFactory.createPseudoErgodicHMM(HMMTrainingParameterSet pars,
double ess,
double selfTranistionPart,
double finalTranistionPart,
AlphabetContainer con,
int numStates,
boolean insertUniform)
Creates an HMM with
numStates+1 states, where numStates emitting build a clique and each of those states is connected to the absorbing silent final state. |
static AbstractHMM |
HMMFactory.createSunflowerHMM(HMMTrainingParameterSet pars,
AlphabetContainer con,
double ess,
int expectedSequenceLength,
boolean startCentral,
int... motifLength)
This method creates a first order sunflower HMM.
|
static AbstractHMM |
HMMFactory.createSunflowerHMM(HMMTrainingParameterSet pars,
AlphabetContainer con,
double ess,
int expectedSequenceLength,
boolean startCentral,
PhyloTree[] t,
double[] motifProb,
int[] motifLength)
This method creates a first order sunflower HMM allowing phylogenetic emissions.
|
Modifier and Type | Method and Description |
---|---|
static Pair<AbstractHMM,HomogeneousMMDiffSM> |
HMMFactory.parseProfileHMMFromHMMer(Reader hmmReader,
StringBuffer consensus,
LinkedList<Integer> matchStates,
LinkedList<Integer> silentStates)
Parses a profile HMM from the textual HMMer representation.
|
Modifier and Type | Class and Description |
---|---|
class |
DifferentiableHigherOrderHMM
This class combines an
HigherOrderHMM and a DifferentiableStatisticalModel by implementing some of the declared methods. |
class |
HigherOrderHMM
This class implements a higher order hidden Markov model.
|
class |
SamplingHigherOrderHMM |
class |
SamplingPhyloHMM
This class implements an (higher order) HMM that contains multi-dimensional emissions described
by a phylogenetic tree.
|