de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous.parameters
Class FSMEMParameterSet

java.lang.Object
  de.jstacs.parameters.ParameterSet
      de.jstacs.parameters.InstanceParameterSet<T>
          de.jstacs.parameters.SequenceScoringParameterSet<T>
              de.jstacs.sequenceScores.statisticalModels.trainable.discrete.DGTrainSMParameterSet
                  de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous.parameters.IDGTrainSMParameterSet
                      de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous.parameters.MEManagerParameterSet
                          de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous.parameters.FSMEMParameterSet

All Implemented Interfaces:

GalaxyConvertible, Storable, Cloneable

public class FSMEMParameterSet
extends MEManagerParameterSet

The ParameterSet for a FSMEManager.

Author:

Jens Keilwagen

See Also:

FSMEManager

Nested Class Summary

Nested classes/interfaces inherited from class de.jstacs.parameters.ParameterSet

ParameterSet.ParameterList

Field Summary

Fields inherited from class de.jstacs.parameters.SequenceScoringParameterSet

alphabet, length

Fields inherited from class de.jstacs.parameters.ParameterSet

errorMessage, parameters, parent

Constructor Summary

FSMEMParameterSet()
The simple constructor.

FSMEMParameterSet(AlphabetContainer alphabet, int length, double ess, String description, ConstraintManager.Decomposition decomposition, boolean reduce, byte algorithm, double epsilon, String constraints)
The fast constructor.

FSMEMParameterSet(StringBuffer s)
The constructor for the Storable interface.

Method Summary

protected void	addParameters() Adds the parameter template in the constructor.
String	getInstanceComment() Returns a comment (a textual description) of the class that can be constructed using this ParameterSet.

Methods inherited from class de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous.parameters.IDGTrainSMParameterSet

getModelInstanceName

Methods inherited from class de.jstacs.sequenceScores.statisticalModels.trainable.discrete.DGTrainSMParameterSet

clone, getInstanceName, hasDefaultOrIsSet, setEss

Methods inherited from class de.jstacs.parameters.SequenceScoringParameterSet

equals, fromXML, getAlphabetContainer, getLength, getNumberOfParameters, getParameterAt, reset, toXML

Methods inherited from class de.jstacs.parameters.InstanceParameterSet

getInstance, getInstanceClass

Methods inherited from class de.jstacs.parameters.ParameterSet

fromGalaxy, getAllParameterNames, getComment, getComment, getErrorMessage, getIndex, getName, getName, getParameterForName, getParent, initParameterList, initParameterList, isAtomic, isComparable, parametersLoaded, setParent, toGalaxy

Methods inherited from class java.lang.Object

finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

FSMEMParameterSet

public FSMEMParameterSet()
                  throws Exception

The simple constructor.

Throws:

Exception - if some parameters templates could not be created properly

FSMEMParameterSet

public FSMEMParameterSet(StringBuffer s)
                  throws NonParsableException

The constructor for the Storable interface.

Parameters:

s - the StringBuffer

Throws:

NonParsableException - if the StringBuffer is not parsable

FSMEMParameterSet

public FSMEMParameterSet(AlphabetContainer alphabet,
                         int length,
                         double ess,
                         String description,
                         ConstraintManager.Decomposition decomposition,
                         boolean reduce,
                         byte algorithm,
                         double epsilon,
                         String constraints)
                  throws Exception

The fast constructor.

Parameters:

alphabet - the alphabet

length - the length of the modeled sequences

ess - the ess

description - the description

decomposition - the kind of decomposition

reduce - whether the constraints should be reduced

algorithm - the choice of algorithm

epsilon - the threshold for stopping the numerical algorithms

constraints - the constraints to be used
There are two ways entering the constraints.

1. Enter single constraints in the following way: node_1, node_2, ..., node_k.
2. Enter groups of constraints in the following way: "m<number of nodes>s<number of skipped nodes>". E.g. m2s1 is the set {0,2; 1,3; 2,4; ...}.
   The "x" can be used to determine that each skip should be used (e.g. m2sx).

To separate the constraints use ";".

Additionally for each node a constraint is assumed (i.e. {0; 1; 2; ...}).

Throws:

Exception - if something went wrong

See Also:

ConstraintManager.Decomposition.DECOMPOSE_NOTHING, ConstraintManager.Decomposition.DECOMPOSE_UNCONNECTED, ConstraintManager.Decomposition.DECOMPOSE_LESS_CONNECTED

Method Detail

addParameters

protected void addParameters()
                      throws Exception

Description copied from class: MEManagerParameterSet

Adds the parameter template in the constructor.

Overrides:

addParameters in class MEManagerParameterSet

Throws:

Exception - if some parameters could not be created properly

getInstanceComment

public String getInstanceComment()

Description copied from class: InstanceParameterSet

Returns a comment (a textual description) of the class that can be constructed using this ParameterSet.

Specified by:

getInstanceComment in class InstanceParameterSet

Returns:

the comment of the class