magpie.models.regression

Class SplitRegression

java.lang.Object

magpie.models.BaseModel

magpie.models.SplitModel

magpie.models.regression.SplitRegression

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, MultiModel, AbstractRegressionModel, Citable, Commandable, Options, Printable, Savable

public class SplitRegression
extends SplitModel
implements AbstractRegressionModel

Abstract class for models that use multiple submodels. Uses robust regression operations copied straight from BaseRegression.

Usage: No options to set.

Implemented Commands:

robust <Q> - Define False Positive Rate used for robust regression. See Motulsky and Brown
Q: Desired FPR during outlier detection.

See Also:

Serialized Form

Field Summary

Fields inherited from class magpie.models.SplitModel

GenericModel, Model, Partitioner

Fields inherited from class magpie.models.BaseModel

AttributeSelector, trained, TrainingStats, validated, ValidationStats

Constructor Summary

Constructors

Constructor and Description
SplitRegression() Create a blank split regression model

Method Summary

Modifier and Type	Method and Description
SplitRegression	clone()
boolean	doRobustRegression() Whether robust regression is desired.
int	getNFittingParameters() Number of fitting parameters in a model.
double	getRobustRegressionQ() Get desired False Discovery Rate
void	robustTraining(Dataset TrainData) Robustly train a model.
java.lang.Object	runCommand(java.util.List<java.lang.Object> Command) Process some command described by a list of Objects.
void	setGenericModel(BaseModel x) Set the model template
void	setModel(int index, BaseModel x) Set a specific submodel.
void	setRobustRegressionQ(double Q) Set the desired False Discovery Rate for outlier detection.

Methods inherited from class magpie.models.SplitModel

checkModelCount, getCitations, getGenericModel, getModel, NModels, printCommand, printModel_protected, printModelDescriptionDetails, printUsage, run_protected, setNumberOfModels, setOptions, setPartitioner, train_protected

Methods inherited from class magpie.models.BaseModel

about, crossValidate, crossValidate, crossValidate, done, externallyValidate, getAttributeSelector, getFilter, getTrainTime, getValidationMethod, handleSetCommand, isTrained, isValidated, loadState, printDescription, printModel, resetModel, run, saveCommand, saveState, setAttributeSelector, setComponent, setFilter, train, train

Methods inherited from class java.lang.Object

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

Constructor Detail

SplitRegression

public SplitRegression()

Create a blank split regression model

Method Detail

clone

public SplitRegression clone()

Overrides:

clone in class SplitModel

setModel

public void setModel(int index,
                     BaseModel x)

Description copied from interface: MultiModel

Set a specific submodel. Will increase number of models if necessary.

Any implementation should not clone the model. This will allow people to construct a model used already-trained models.

Specified by:

setModel in interface MultiModel

Overrides:

setModel in class SplitModel

Parameters:

index - Index of submodel to be set

x - Model to be used (creates a clone)

setGenericModel

public void setGenericModel(BaseModel x)

Description copied from class: SplitModel

Set the model template

Specified by:

setGenericModel in interface MultiModel

Overrides:

setGenericModel in class SplitModel

Parameters:

x - Template model (will be cloned)

robustTraining

public void robustTraining(Dataset TrainData)

Description copied from interface: AbstractRegressionModel

Robustly train a model. Is called from BaseModel.train(magpie.data.Dataset).

I recommend that you use methods outlined in a paper by Motulsky and Brown.

.

Specified by:

robustTraining in interface AbstractRegressionModel

Parameters:

TrainData - Training data

setRobustRegressionQ

public void setRobustRegressionQ(double Q)

Description copied from interface: AbstractRegressionModel

Set the desired False Discovery Rate for outlier detection. A robust regression parameter of 0.01 is recommended. Zero turns it off. Don't give it less than zero.

See paper by Benjamini and Hochberg

Specified by:

setRobustRegressionQ in interface AbstractRegressionModel

Parameters:

Q - Desired FDR

getRobustRegressionQ

public double getRobustRegressionQ()

Description copied from interface: AbstractRegressionModel

Get desired False Discovery Rate

Specified by:

getRobustRegressionQ in interface AbstractRegressionModel

Returns:

Target FDR

doRobustRegression

public boolean doRobustRegression()

Description copied from interface: AbstractRegressionModel

Whether robust regression is desired. For now, this should be:
getRobustRegressionQ > 0

Specified by:

doRobustRegression in interface AbstractRegressionModel

getNFittingParameters

public int getNFittingParameters()

Description copied from interface: AbstractRegressionModel

Number of fitting parameters in a model. Only pertinent to nonlinear regression. Otherwise, this should just return 0.

Specified by:

getNFittingParameters in interface AbstractRegressionModel

Returns:

Number of fitting parameters

runCommand

public java.lang.Object runCommand(java.util.List<java.lang.Object> Command)
                            throws java.lang.Exception

Description copied from interface: Commandable

Process some command described by a list of Objects.

Specified by:

runCommand in interface Commandable

Overrides:

runCommand in class SplitModel

Parameters:

Command - Command as a list of objects

Returns:

Any output (null if no output generated by command)

Throws:

java.lang.Exception - If something goes wrong