de.jstacs.models
Class AbstractModel

java.lang.Object
  de.jstacs.models.AbstractModel

All Implemented Interfaces:

Model, Storable, Cloneable

Direct Known Subclasses:

AbstractMixtureModel, CompositeModel, DiscreteGraphicalModel, UniformModel

public abstract class AbstractModel
extends Object
implements Cloneable, Storable, Model

Abstract class for a model for pattern recognition.
For writing or reading a StringBuffer to or from a file (fromXML(StringBuffer), toXML()) you can use the class FileManager.

Author:

Andre Gohr, Jan Grau, Jens Keilwagen

See Also:

FileManager

Field Summary

protected AlphabetContainer	alphabets The underlying alphabets
protected int	length the length of the sequences the model can classify.

Constructor Summary

AbstractModel(AlphabetContainer alphabets, int length)
Constructor that sets the length of the model to length and the AlphabetContainer to alphabets.

AbstractModel(StringBuffer stringBuff)
Constructor for parsing a AbstractModel out of a StringBuffer.

Method Summary

AbstractModel	clone() Follows the conventions of Object's clone-method.
Sample	emitSample(int numberOfSequences, int... seqLength) This method returns a Sample object containing artificial sequence(s).
protected abstract void	fromXML(StringBuffer xml) This method should only be used by the constructor that works on StringBuffer.
AlphabetContainer	getAlphabetContainer() Returns the container of alphabets that were used when constructing the model.
ResultSet	getCharacteristics() Returns some information characterizing or describing the current instance of the model.
int	getLength() Returns the length of sequence this model can classify.
double[]	getLogProbFor(Sample data) This method computes the logarithm of the probabilities of all sequences in the given sample.
void	getLogProbFor(Sample data, double[] res) This method computes and stores the logarithm of the probabilities for any sequence in the sample in the given double array.
double	getLogProbFor(Sequence sequence) Returns the logarithm of the probability of the given sequence given the model.
double	getLogProbFor(Sequence sequence, int startpos) Returns the logarithm of the probability of the given sequence given the model.
double	getLogProbFor(Sequence sequence, int startpos, int endpos) Returns the logarithm of the probability of the given sequence given the model.
byte	getMaximalMarkovOrder() This method returns the maximal used markov order if possible.
double	getPriorTerm() Returns a value that is proportional to the prior.
double	getProbFor(Sequence sequence) Returns the probability of the given sequence given the model.
double	getProbFor(Sequence sequence, int startpos) Returns the probability of the given sequence given the model.
protected void	set(AlphabetContainer abc) This method should only be invoked by the method setNewAlphabetContainerInstance( AlphabetContainer ) and not be made public.
boolean	setNewAlphabetContainerInstance(AlphabetContainer abc) This method tries to set a new instance of an AlphabetContainer for the current model.
void	train(Sample data) Trains the AbstractModel object given the data as Sample.

Methods inherited from class java.lang.Object

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

Methods inherited from interface de.jstacs.models.Model

getInstanceName, getLogPriorTerm, getNumericalCharacteristics, getProbFor, isTrained, toString, train

Field Detail

length

protected int length

the length of the sequences the model can classify. For models that can take sequences of arbitrary length this should be set to 0

alphabets

protected AlphabetContainer alphabets

The underlying alphabets

Constructor Detail

AbstractModel

public AbstractModel(AlphabetContainer alphabets,
                     int length)

Constructor that sets the length of the model to length and the AlphabetContainer to alphabets.
The parameter length gives the length of the sequences the model can classify. Models that can only classify sequences of defined length are e.g. PWM or inhomogeneous Markov models. If the model can classify sequences of arbitrary length, e.g. homogeneous Markov models, this parameter must be set to 0 (zero).
The length and alphabets define the type of data that can be modeled and therefor both has to be checked before any evaluation (e.g. getProbFor)

Parameters:

alphabets - the alphabets

length - the length of the sequences a model can classify, 0 for arbitrary length

AbstractModel

public AbstractModel(StringBuffer stringBuff)
              throws NonParsableException

Constructor for parsing a AbstractModel out of a StringBuffer.

Parameters:

stringBuff - the StringBuffer to be parsed

Throws:

NonParsableException - is thrown if the StringBuffer could not be parsed

Method Detail

clone

public AbstractModel clone()
                    throws CloneNotSupportedException

Follows the conventions of Object's clone-method.

Specified by:

clone in interface Model

Overrides:

clone in class Object

Returns:

an object, that is a copy of the current AbstractModel (the member-AlphabetContainer isn't deeply cloned since it is assumed to be immutable). The type of the returned object is defined by the class X directly inherited from AbstractModel. Hence X's clone-method should work as:
1. Object o = (X)super.clone(); 2. all additional member variables of o defined by X that are not of simple data-types like int, double, ... , have to be deeply copied 3. return o

Throws:

CloneNotSupportedException

train

public void train(Sample data)
           throws Exception

Description copied from interface: Model

Trains the AbstractModel object given the data as Sample.
This method should work non-incrementally. That means the result of the following series: train(data1); train(data2) should be a fully trained model over data2 and not over data1+data2. All parameters of the model were given by the call of the constructor.

Specified by:

train in interface Model

Parameters:

data - the given sequences as Sample

Throws:

Exception - an Exception should be thrown if the training did not succeed.

See Also:

Sample.getElementAt(int), Sample.ElementEnumerator

getProbFor

public double getProbFor(Sequence sequence)
                  throws NotTrainedException,
                         Exception

Description copied from interface: Model

Returns the probability of the given sequence given the model. If a least one random variable is continuous the value of density function is returned.
The length and alphabets define the type of data that can be modeled and therefore both has to be checked.

Specified by:

getProbFor in interface Model

Parameters:

sequence - the sequence

Returns:

the probability or the value of the density function of the given sequence given the model

Throws:

NotTrainedException - a NotTrainedException should be thrown if the model is not trained yet.

Exception - an Exception should be thrown if the sequence could not be handled by the model

getProbFor

public double getProbFor(Sequence sequence,
                         int startpos)
                  throws NotTrainedException,
                         Exception

Description copied from interface: Model

Returns the probability of the given sequence given the model. If a least one random variable is continuous the value of density function is returned.

If the length of the sequences, whose probability should be returned, is fixed (e.g. in a inhomogenous model) and the given sequence is longer than their fixed length, the start position within the given sequence is given by startpos. E.g. the fixed length is 12. The length of the given sequence is 30 and the startpos=15 the probability of the part from position 15 to 26 (inclusive) given the model should be returned.
The length and alphabets define the type of data that can be modeled and therefore both has to be checked.

Specified by:

getProbFor in interface Model

Parameters:

sequence - the sequence

startpos - the start position

Returns:

the probability or the value of the density function of the part of the given sequence given the model

Throws:

NotTrainedException - a NotTrainedException should be thrown if the model is not trained yet.

Exception - an Exception should be thrown if the sequence could not be handled by the model

getLogProbFor

public double getLogProbFor(Sequence sequence,
                            int startpos,
                            int endpos)
                     throws Exception

Description copied from interface: Model

Returns the logarithm of the probability of the given sequence given the model. If a least one random variable is continuous the value of density function is returned.

For more details see Model.getProbFor(Sequence, int, int)

Specified by:

getLogProbFor in interface Model

Parameters:

sequence - the sequence

startpos - the start position

endpos - the last position to be taken into account

Returns:

the logarithm of probability or the value of the density function of the part of the given sequence given the model

Throws:

Exception - an Exception should be thrown if the sequence could not be handled (e.g. startpos > endpos, endpos > sequence.length, ...) by the model

NotTrainedException - a NotTrainedException should be thrown if the model is not trained yet.

See Also:

Model.getProbFor(Sequence, int, int)

getLogProbFor

public double getLogProbFor(Sequence sequence,
                            int startpos)
                     throws Exception

Description copied from interface: Model

Returns the logarithm of the probability of the given sequence given the model. If a least one random variable is continuous the value of density function is returned.

For more details see Model.getProbFor(Sequence, int)

Specified by:

getLogProbFor in interface Model

Parameters:

sequence - the sequence

startpos - the start position

Returns:

the logarithm of probability or the value of the density function of the part of the given sequence given the model

Throws:

Exception - an Exception should be thrown if the sequence could not be handled by the model

NotTrainedException - a NotTrainedException should be thrown if the model is not trained yet.

See Also:

Model.getProbFor(Sequence, int)

getLogProbFor

public double getLogProbFor(Sequence sequence)
                     throws Exception

Description copied from interface: Model

Returns the logarithm of the probability of the given sequence given the model. If a least one random variable is continuous the value of density function is returned.

For more details see Model.getProbFor(Sequence)

Specified by:

getLogProbFor in interface Model

Parameters:

sequence - the sequence

Returns:

the logarithm of probability or the value of the density function of the part of the given sequence given the model

Throws:

Exception - an Exception should be thrown if the sequence could not be handled by the model

NotTrainedException - a NotTrainedException should be thrown if the model is not trained yet.

See Also:

Model.getProbFor(Sequence)

getLogProbFor

public double[] getLogProbFor(Sample data)
                       throws Exception

Description copied from interface: Model

This method computes the logarithm of the probabilities of all sequences in the given sample. The values are stored in the array according to the index of the sequence in the sample.

The probability for any sequence shall be computed independent of all other sequences in the sample. So the result should be exactly the same as for the method getLogProbFor(Sequence)

Specified by:

getLogProbFor in interface Model

Parameters:

data - the sample

Returns:

an array containing the logarithm of the probabilities of all sequences of the sample

Throws:

Exception - if something went wrong

See Also:

Model.getLogProbFor(Sequence)

getLogProbFor

public void getLogProbFor(Sample data,
                          double[] res)
                   throws Exception

Description copied from interface: Model

This method computes and stores the logarithm of the probabilities for any sequence in the sample in the given double array.

The probability for any sequence shall be computed independent of all other sequences in the sample. So the result should be exactly the same as for the method getLogProbFor(Sequence)

Specified by:

getLogProbFor in interface Model

Parameters:

data - the sample

res - the array for the results, has to have length data.getNumberOfElements()

Throws:

Exception - if something went wrong

See Also:

Model.getLogProbFor(Sample)

getPriorTerm

public double getPriorTerm()
                    throws Exception

Description copied from interface: Model

Returns a value that is proportional to the prior. For ML 1 should be returned.

Specified by:

getPriorTerm in interface Model

Returns:

a value that is proportional to the prior

Throws:

Exception - if something went wrong

emitSample

public Sample emitSample(int numberOfSequences,
                         int... seqLength)
                  throws NotTrainedException,
                         Exception

Description copied from interface: Model

This method returns a Sample object containing artificial sequence(s).

There are 2 different possibilities to create a sample for a model with length 0.

1. emitSample( int n, int l ) should return a sample with n sequences of length l.
2. emitSample( int n, int[] l ) should return a sample with n sequences which have a sequence length corresponding to the entry in the array

There are 2 different possibilities to create a sample for a model with length greater than 0. emitSample( int n ) and emitSample( int n, null ) should return a sample with n sequences of length of the model (Model.getLength())

The standard implementation throws an Exception.

Specified by:

emitSample in interface Model

Parameters:

numberOfSequences - the number of sequences that should be contained in the returned sample

seqLength - the length of the sequences for a homogeneous model; for an inhomogeneous model this parameter should be null or an array of size 0.

Returns:

Sample containing the artificial sequence(s)

Throws:

NotTrainedException - a NotTrainedException should be thrown if the model is not trained yet.

Exception - an Exception should be thrown if the emission did not succeed.

See Also:

Sample

getAlphabetContainer

public final AlphabetContainer getAlphabetContainer()

Description copied from interface: Model

Returns the container of alphabets that were used when constructing the model.

Specified by:

getAlphabetContainer in interface Model

Returns:

the alphabet

getLength

public final int getLength()

Description copied from interface: Model

Returns the length of sequence this model can classify. Models that can only classify sequences of defined length are e.g. PWM or inhomogeneous Markov models. If the model can classify sequences of arbitrary length, e.g. homogeneous Markov models, this method returns 0 (zero).

Specified by:

getLength in interface Model

Returns:

the length

getMaximalMarkovOrder

public byte getMaximalMarkovOrder()
                           throws UnsupportedOperationException

Description copied from interface: Model

This method returns the maximal used markov order if possible.

Specified by:

getMaximalMarkovOrder in interface Model

Returns:

maximal used markov order

Throws:

UnsupportedOperationException - if the model can't give a proper answer

getCharacteristics

public ResultSet getCharacteristics()
                             throws Exception

Description copied from interface: Model

Returns some information characterizing or describing the current instance of the model. This could be e.g. the number of edges for a Bayesian network or an image showing some representation of the model. The set of characteristics should always include the XML-representation of the model. The corresponding result type is ObjectResult

Specified by:

getCharacteristics in interface Model

Returns:

the characteristics

Throws:

Exception - an Exception is thrown if some of the characteristics could not be defined

See Also:

StorableResult

fromXML

protected abstract void fromXML(StringBuffer xml)
                         throws NonParsableException

This method should only be used by the constructor that works on StringBuffer. It is the counter part of toXML().

Parameters:

xml - the representation

Throws:

NonParsableException - if the StringBuffer is not parsable or the representation is conflicting

See Also:

AbstractModel(StringBuffer)

setNewAlphabetContainerInstance

public final boolean setNewAlphabetContainerInstance(AlphabetContainer abc)

Description copied from interface: Model

This method tries to set a new instance of an AlphabetContainer for the current model. This instance has to be consistent with the underlying instance of an AlphabetContainer.

This method can be very usefull to save time.

Specified by:

setNewAlphabetContainerInstance in interface Model

Parameters:

abc - the alphabets

Returns:

true if the new instance could be set

See Also:

Model.getAlphabetContainer(), AlphabetContainer.checkConsistency(AlphabetContainer)

set

protected void set(AlphabetContainer abc)

This method should only be invoked by the method setNewAlphabetContainerInstance( AlphabetContainer ) and not be made public.

It enables you to do more with the method setNewAlphabetContainerInstance( AlphabetContainer ), e.g. setting a new AlphabetContainer instance for subcomponents.

Parameters:

abc - the new instance