|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Objectde.jstacs.scoringFunctions.AbstractNormalizableScoringFunction
de.jstacs.scoringFunctions.directedGraphicalModels.BayesianNetworkScoringFunction
public class BayesianNetworkScoringFunction
This class implements a scoring function that is a moral directed graphical model, i.e. a moral Bayesian network.
This implementation also comprises well known specializations of Bayesian networks like Markov models of arbitrary order
(including weight array matrix models and position weight matrices) or Bayesian trees. Different structures can be achieved by using the
corresponding Measure
, e.g. InhomogeneousMarkov
for Markov models of arbitrary order.
This scoring function can be used in any ScoreClassifier
, e.g. in a CLLClassifier
to learn the parameters of the ScoringFunction
using maximum conditional likelihood.
Field Summary | |
---|---|
protected double |
ess
The equivalent sample size |
protected boolean |
isTrained
Indicates if the instance has been trained |
protected Double |
normalizationConstant
Normalization constant to obtain normalized probabilities |
protected Integer |
numFreePars
The number of free parameters. |
protected int[] |
nums
Used, internally. |
protected int[][] |
order
network structure, used internally |
protected Parameter[] |
parameters
The parameters of the scoring function. |
protected boolean |
plugInParameters
Indicates if plug-in parameters, i.e. generative (MAP) parameters shall be used upon initialization |
protected Measure |
structureMeasure
Measure that defines the network structure |
protected ParameterTree[] |
trees
The trees that represent the context of the random variable (i.e. configuration of parent random variables) of the parameters. |
Fields inherited from class de.jstacs.scoringFunctions.AbstractNormalizableScoringFunction |
---|
alphabets, length, r |
Fields inherited from interface de.jstacs.scoringFunctions.ScoringFunction |
---|
UNKNOWN |
Constructor Summary | |
---|---|
BayesianNetworkScoringFunction(AlphabetContainer alphabet,
int length,
double ess,
boolean plugInParameters,
Measure structureMeasure)
Creates a new BayesianNetworkScoringFunction that has neither been initialized nor trained. |
|
BayesianNetworkScoringFunction(StringBuffer xml)
Re-creates a BayesianNetworkScoringFunction from its XML-representation, as saved by the toXML() } method. |
Method Summary | |
---|---|
void |
addGradientOfLogPriorTerm(double[] grad,
int start)
This method computes the gradient of getLogPriorTerm() for each parameter of this model. |
BayesianNetworkScoringFunction |
clone()
Creates a clone (deep copy) of the current ScoringFunction instance. |
protected void |
createTrees(Sample[] data2,
double[][] weights2)
Creates the tree structures that represent the context (array trees )
and the parameter objects parameters using the given Measure structureMeasure . |
protected void |
fromXML(StringBuffer source)
This method is called in the constructor to create a scoring function from a StringBuffer |
double[] |
getCurrentParameterValues()
Returns a double array of dimension getNumberOfParameters() containing the current parameter
values. |
double |
getEss()
Returns the equivalent sample size of this model, i.e. the equivalent sample size for the class or component that is represented by this model. |
String |
getInstanceName()
Returns a short instance name. |
double |
getLogPriorTerm()
This method computes a value that is proportional to getESS()*Math.log( getNormalizationConstant() ) + Math.log( prior ). |
double |
getLogScore(Sequence seq,
int start)
Returns the log score for the sequence |
double |
getLogScoreAndPartialDerivation(Sequence seq,
int start,
IntList indices,
DoubleList partialDer)
Returns the log score for the sequence and fills the list with the indices and the partial derivations. |
double |
getNormalizationConstant()
Returns the sum of the scores over all sequences of the event space. |
int |
getNumberOfParameters()
The number of parameters in this scoring function. |
double |
getPartialNormalizationConstant(int parameterIndex)
Returns the partial normalization constant for the parameter with index parameterIndex . |
int |
getPositionForParameter(int index)
Returns the position in the sequence, the parameter index is responsible for. |
double[][] |
getPWM()
If this BayesianNetworkScoringFunction is a PWM, i.e. |
int |
getSizeOfEventSpaceForRandomVariablesOfParameter(int index)
Returns the size of the event space of the random variables that are affected by parameter no. |
void |
initializeFunction(int index,
boolean freeParams,
Sample[] data,
double[][] weights)
This method creates the underlying structure of the scoring function. |
void |
initializeFunctionRandomly(boolean freeParams)
This method initializes the scoring function randomly. |
boolean |
isInitialized()
This method can be used to determine whether the model is initialized. |
protected void |
precomputeNormalization()
Precomputes all normalization constants and saves the global normalization constant to normalizationConstant . |
void |
setParameters(double[] params,
int start)
This method sets the internal parameters to the values of params between start and
start + this.getNumberOfParameters() - 1 |
protected void |
setPlugInParameters(int index,
boolean freeParameters,
Sample[] data,
double[][] weights)
Computes and sets the plug-in parameters (MAP estimated parameters) from data using weights . |
String |
toString()
|
StringBuffer |
toXML()
This method returns an XML-representation of an instance of the implementing class. |
Methods inherited from class de.jstacs.scoringFunctions.AbstractNormalizableScoringFunction |
---|
getAlphabetContainer, getInitialClassParam, getLength, getLogScore, getLogScoreAndPartialDerivation, getNumberOfRecommendedStarts, isNormalized, isNormalized |
Methods inherited from class java.lang.Object |
---|
equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait |
Field Detail |
---|
protected Parameter[] parameters
protected ParameterTree[] trees
protected boolean isTrained
protected double ess
protected Integer numFreePars
protected int[] nums
protected Measure structureMeasure
Measure
that defines the network structure
protected boolean plugInParameters
protected int[][] order
protected Double normalizationConstant
Constructor Detail |
---|
public BayesianNetworkScoringFunction(AlphabetContainer alphabet, int length, double ess, boolean plugInParameters, Measure structureMeasure) throws Exception
alphabet
- the alphabet of the scoring function boxed in an AlphabetContainer
, e.g new AlphabetContainer(new DNAAlphabet
())
length
- the length of the scoring function, i.e. the length of the sequences this scoring function can handleess
- the equivalent sample sizeplugInParameters
- indicates if plug-in parameters, i.e. generative (MAP) parameters shall be used upon initialzationstructureMeasure
- the Measure
used for the structure, e.g. InhomogeneousMarkov
Exception
- an exception is thrown if the length of the scoring function is not admissible (<=0), or the alphabet is not discretepublic BayesianNetworkScoringFunction(StringBuffer xml) throws NonParsableException
toXML()
} method.
xml
- the XML-representation
NonParsableException
- if the XML-code could not be parsed an exception is thrownMethod Detail |
---|
public BayesianNetworkScoringFunction clone() throws CloneNotSupportedException
ScoringFunction
ScoringFunction
instance.
clone
in interface ScoringFunction
clone
in class AbstractNormalizableScoringFunction
CloneNotSupportedException
public double getPartialNormalizationConstant(int parameterIndex) throws Exception
NormalizableScoringFunction
parameterIndex
. This is
the partial derivation of the normalization constant for the parameter with index parameterIndex
\frac{\partial Z(\lambda)}{\partial \lambda_{index}}
.
parameterIndex
- the index of the parameter
Exception
- if something went wrong with the Normalizationpublic void initializeFunction(int index, boolean freeParams, Sample[] data, double[][] weights) throws Exception
ScoringFunction
index
- the index of the class the scoring function modelsfreeParams
- if true, the (reduced) parameterization is useddata
- the samplesweights
- the weights of the sequences in the samples
- Throws:
Exception
protected void createTrees(Sample[] data2, double[][] weights2) throws Exception
trees
)
and the parameter objects parameters
using the given Measure
structureMeasure
.
data2
- the data that is used to compute the structureweights2
- the weights on the sequences in data2
Exception
- throws an Exception if the structure is no moral graph or if the lengths of data and scoring function do not match or other problems concerning the data occurprotected void setPlugInParameters(int index, boolean freeParameters, Sample[] data, double[][] weights)
data
using weights
.
index
- the index of the class the scoring function is responsible for. The parameters are estimated from data[index]
and weights[index]
.freeParameters
- indicates if only the free parameters or all parameters should be used. This also affects the initialization.data
- the data used for initializationweights
- the weights on the dataprotected void fromXML(StringBuffer source) throws NonParsableException
AbstractNormalizableScoringFunction
fromXML
in class AbstractNormalizableScoringFunction
source
- the XML representation
NonParsableException
- if the StringBuffer could not be parsed.public String toString()
toString
in class Object
public String getInstanceName()
ScoringFunction
public double getLogScore(Sequence seq, int start)
ScoringFunction
seq
- the sequencestart
- the startposition in the sequence
public double getLogScoreAndPartialDerivation(Sequence seq, int start, IntList indices, DoubleList partialDer)
ScoringFunction
seq
- the sequencestart
- the startposition in the sequenceindices
- after method invocation the list should contain the indices i where \frac{\partial \log
score(seq)}{\partial \lambda_i} is not zeropartialDer
- after method invocation the list should contain the corresponding \frac{\partial \log
score(seq)}{\partial \lambda_i}
public double getNormalizationConstant() throws RuntimeException
NormalizableScoringFunction
RuntimeException
public int getNumberOfParameters()
ScoringFunction
UNKNOWN
.
ScoringFunction.UNKNOWN
public void setParameters(double[] params, int start)
ScoringFunction
params
between start
and
start + this.getNumberOfParameters() - 1
params
- the parametersstart
- the start indexprotected void precomputeNormalization()
normalizationConstant
.
RuntimeException
public double[] getCurrentParameterValues() throws Exception
ScoringFunction
getNumberOfParameters()
containing the current parameter
values. If on e likes to use these parameters to start an optimization it is highly recommended to invoke
ScoringFunction.initializeFunction(int, boolean, Sample[], double[][])
before. After an optimization
this method can be used to get the current parameter values.
Exception
- is thrown if no parameters exist, yetpublic StringBuffer toXML()
Storable
public double getLogPriorTerm()
NormalizableScoringFunction
getESS()*Math.log( getNormalizationConstant() ) + Math.log( prior ).
prior is the prior for the parameters of this model.
- Returns:
getESS()*Math.log( getNormalizationConstant() ) + Math.log( prior )
- See Also:
NormalizableScoringFunction.getEss()
,
NormalizableScoringFunction.getNormalizationConstant()
public void addGradientOfLogPriorTerm(double[] grad, int start)
NormalizableScoringFunction
getLogPriorTerm()
for each parameter of this model. The
results are added to the array grad
beginning at index start
.
grad
- the gradientstart
- the start index in the grad
array, where the partial derivations for the parameters of
this models shall be enterNormalizableScoringFunction.getLogPriorTerm()
public double getEss()
NormalizableScoringFunction
public int getPositionForParameter(int index)
index
is responsible for.
index
- the index of the parameter
public int getSizeOfEventSpaceForRandomVariablesOfParameter(int index)
NormalizableScoringFunction
index
,
i.e. the product of the sizes of the alphabets at the position of each random variable affected by parameter
index
. For DNA-alphabets this corresponds to 4 for a PWM, 16 for a WAM except position 0, ...
index
- the index of the parameter
public void initializeFunctionRandomly(boolean freeParams) throws Exception
ScoringFunction
freeParams
- if true, the (reduced) parameterization is used
Exception
public boolean isInitialized()
ScoringFunction
ScoringFunction.initializeFunction(int, boolean, Sample[], double[][])
.
true
if the model is initializedpublic double[][] getPWM() throws Exception
BayesianNetworkScoringFunction
is a PWM, i.e. structureMeasure
=new InhomogeneousMarkov
(0)}},
this method returns the normalized PWM as a double array of dimension AbstractNormalizableScoringFunction.getLength()
x size-of-alphabet
Exception
- throws an Exception if this method is called for a BayesianNetworkScoringFunction
} that is not a PWM
|
||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |