de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous

Class MEM

java.lang.Object

de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous.MEM

All Implemented Interfaces:

Storable, Cloneable

public final class MEM
extends Object
implements Storable, Cloneable

This class represents a maximum entropy model. It should not be used without the MEManager.

Author:

Jens Keilwagen

Field Summary

Fields

Modifier and Type	Field and Description
protected MEMConstraint[]	constraints The specific constraints of this MEM.

Constructor Summary

Constructors

Constructor and Description
MEM(AbstractList<int[]> constr, int[] allAlphLen, int[] indices) The main constructor of a MEM.
MEM(int[] constraint, int[] allAlphLen, int[][] cond) The main constructor of a MEM.
MEM(StringBuffer representation) The constructor for the Storable interface.

Method Summary

Methods

Modifier and Type	Method and Description
MEM	clone()
double	getLogPriorPart(double ess) This method compute the prior for the current parameter ignoring some constants.
double	getLogScoreFor(Sequence seq, int start) Returns the logarithmic score for the sequence beginning at start.
double	getScoreFor(Sequence seq, int start) Returns the score for the sequence beginning at start.
String	toString(NumberFormat nf) This method returns some String using a given NumberFormat.
StringBuffer	toXML() This method returns an XML representation as StringBuffer of an instance of the implementing class.
void	train(SequenceIterator s, byte algo, TerminationCondition condition, SafeOutputStream sostream) This method approximates the distribution either analytically or numerically.

Methods inherited from class java.lang.Object

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

Field Detail

constraints

protected MEMConstraint[] constraints

The specific constraints of this MEM.

Constructor Detail

MEM

public MEM(AbstractList<int[]> constr,
   int[] allAlphLen,
   int[] indices)

The main constructor of a MEM.

Parameters:

constr - the constraints

allAlphLen - the alphabet length of all alphabets

indices - the indices of the positions to be used

MEM

public MEM(int[] constraint,
   int[] allAlphLen,
   int[][] cond)

The main constructor of a MEM.

Parameters:

constraint - the constraint

allAlphLen - the alphabet length of all alphabets

cond - the conditional fixed positions

MEM

public MEM(StringBuffer representation)
    throws NonParsableException

The constructor for the Storable interface.

Parameters:

representation - the representation

Throws:

NonParsableException - if the the StringBuffer could not be parsed.

Method Detail

clone

public MEM clone()
          throws CloneNotSupportedException

Overrides:

clone in class Object

Throws:

CloneNotSupportedException

getScoreFor

public double getScoreFor(Sequence seq,
                 int start)

Returns the score for the sequence beginning at start.

Parameters:

seq - the sequence

start - the start position

Returns:

the score for the sequence beginning at start

getLogScoreFor

public double getLogScoreFor(Sequence seq,
                    int start)

Returns the logarithmic score for the sequence beginning at start.

Parameters:

seq - the sequence

start - the start position

Returns:

the logarithmic score for the sequence beginning at start

getLogPriorPart

public double getLogPriorPart(double ess)

This method compute the prior for the current parameter ignoring some constants.

Parameters:

ess - the ESS to be used

Returns:

the prior for the current parameter ignoring some constants

See Also:

StatisticalModel.getLogPriorTerm()

toString

public String toString(NumberFormat nf)

This method returns some String using a given NumberFormat.

Parameters:

nf - the NumberFormat.

Returns:

some String representation

toXML

public StringBuffer toXML()

Description copied from interface: Storable

This method returns an XML representation as StringBuffer of an instance of the implementing class.

Specified by:

toXML in interface Storable

Returns:

the XML representation

train

public void train(SequenceIterator s,
         byte algo,
         TerminationCondition condition,
         SafeOutputStream sostream)
           throws Exception

This method approximates the distribution either analytically or numerically.

Parameters:

s - the SequenceIterator used in normalization and numerical approximation

algo - the choice of numerical approximation

condition - the TerminationCondition for stopping the iterative algorithm

sostream - a possibility for writing some information

Throws:

Exception - if something went wrong inside the algorithms

See Also:

MEMTools