jsat.datatransform

Class DataTransformProcess

java.lang.Object

jsat.datatransform.DataTransformProcess

All Implemented Interfaces:

Serializable, Cloneable, DataTransform, Parameterized

public class DataTransformProcess
extends Object
implements DataTransform, Parameterized

Performing a transform on the whole data set before training a classifier can add bias to the results. For proper evaluation, the transforms must be learned from the training set and not contain any knowledge from the testing set. A DataTransformProcess aids in this by providing a mechanism to contain several different transforms to learn and then apply.

The Parameters of the Data Transform Process are the parameters from the individual transform factories that make up the whole process. The name "DataTransformProcess" will not be prefixed to the parameter names.

Author:

Edward Raff

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
DataTransformProcess() Creates a new transform process that is empty.
DataTransformProcess(DataTransform... transforms) Creates a new transform process from the listed factories, which will be applied in order by index.

Method Summary

Modifier and Type	Method and Description
void	addTransform(DataTransform transform) Adds a transform to the list of transforms.
DataTransformProcess	clone()
void	fit(DataSet data) Fits this transform to the given dataset.
int	getNumberOfTransforms()
Parameter	getParameter(String paramName) Returns the parameter with the given name.
List<Parameter>	getParameters() Returns the list of parameters that can be altered for this learner.
void	leanTransforms(DataSet dataSet) Learns the transforms for the given data set.
void	learnApplyTransforms(DataSet dataSet) Learns the transforms for the given data set.
DataPoint	transform(DataPoint dp) Returns a new data point that is a transformation of the original data point.

Methods inherited from class java.lang.Object

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

Constructor Detail

DataTransformProcess

public DataTransformProcess()

Creates a new transform process that is empty. Transform factories must be added using #addTransform(jsat.datatransform.DataTransformFactory) .

DataTransformProcess

public DataTransformProcess(DataTransform... transforms)

Creates a new transform process from the listed factories, which will be applied in order by index.

Parameters:

transforms - the array of factories to apply as the data transform process

Method Detail

addTransform

public void addTransform(DataTransform transform)

Adds a transform to the list of transforms. Transforms are learned and applied in the order in which they are added.

Parameters:

transform - the factory for the transform to add

getNumberOfTransforms

public int getNumberOfTransforms()

Returns:

the number of transforms currently chained in this transform process

fit

public void fit(DataSet data)

Description copied from interface: DataTransform

Fits this transform to the given dataset. Some transforms can only be learned from classification or regression datasets. If an incompatible dataset type is given, a FailedToFitException exception may be thrown.

Specified by:

fit in interface DataTransform

Parameters:

data - the dataset to fir this transform to

leanTransforms

public void leanTransforms(DataSet dataSet)

Learns the transforms for the given data set. The data set will not be altered. Once finished, this DataTransformProcess can be applied to the dataSet to get the transformed data set.

Parameters:

dataSet - the data set to learn a series of transforms from

learnApplyTransforms

public void learnApplyTransforms(DataSet dataSet)

Learns the transforms for the given data set. The data set is then altered after each transform is learned so the next transform can be learned as well.
The results are equivalent to calling learnApplyTransforms(jsat.DataSet) on the data set and then calling DataSet.applyTransform(jsat.datatransform.DataTransform) with this DataTransformProces.

Parameters:

dataSet - the data set to learn a series of transforms from and alter into the final transformed form

transform

public DataPoint transform(DataPoint dp)

Description copied from interface: DataTransform

Returns a new data point that is a transformation of the original data point. This new data point is a different object, but may contain the same references as the original data point. It is not guaranteed that you can mutate the transformed point without having a side effect on the original point.

Specified by:

transform in interface DataTransform

Parameters:

dp - the data point to apply a transformation to

Returns:

a transformed data point

clone

public DataTransformProcess clone()

Specified by:

clone in interface DataTransform

Overrides:

clone in class Object

getParameters

public List<Parameter> getParameters()

Description copied from interface: Parameterized

Returns the list of parameters that can be altered for this learner.

Specified by:

getParameters in interface Parameterized

Returns:

the list of parameters that can be altered for this learner.

getParameter

public Parameter getParameter(String paramName)

Description copied from interface: Parameterized

Returns the parameter with the given name. Two different strings may map to a single Parameter object. An ASCII only string, and a Unicode style string.

Specified by:

getParameter in interface Parameterized

Parameters:

paramName - the name of the parameter to obtain

Returns:

the Parameter in question, or null if no such named Parameter exists.