com.emphysic.myriad.core.data.ops

Class FeatureScalingOperation

java.lang.Object

com.emphysic.myriad.core.data.ops.FeatureScalingOperation

All Implemented Interfaces:

DatasetOperation, ObjectMap, com.esotericsoftware.kryo.KryoSerializable

public class FeatureScalingOperation
extends java.lang.Object
implements DatasetOperation

FeatureScalingOperation - scales features for machine learning by subtracting the mean and dividing by the standard deviation for each feature (index position).

Constructor Summary

Constructors

Constructor and Description
FeatureScalingOperation() Creates a new FeatureScalingOperation.
FeatureScalingOperation(java.util.List<? extends Dataset> datasets) Creates a new scaler and fits with the provided data.

Method Summary

Modifier and Type	Method and Description
void	fit(java.util.List<? extends Dataset> datasets) "Trains" (i.e.
java.util.List<java.lang.Double>	getFeatureMeans() Returns the feature means
java.util.List<java.lang.Double>	getFeatureStds() Returns the feature standard deviations
java.util.Map<java.lang.String,java.lang.Object>	getObjectMap() Creates a map of the important fields for the instance, suitable for serialization.
OnlineStats	getOnlineStats() Returns the online statistics approximator.
long	getSerializationVersion() Returns the current version of the serialization format.
int	getVersion() Returns the current class version.
void	initCurrentVersion(java.util.Map<java.lang.String,java.lang.Object> objectMap) Initializes an instance with a current-version object graph.
static void	main(java.lang.String[] args)
void	partial_fit(java.util.List<? extends Dataset> datasets) Incremental training - keeps approximations of feature means and standard deviations.
Dataset	run(Dataset input) Scales the input data by substracting the feature means and dividing by their standard deviations.
void	setFeatureMeans(java.util.List<java.lang.Double> featureMeans) Sets the feature means
void	setFeatureStds(java.util.List<java.lang.Double> featureStds) Sets the feature standard deviations

Methods inherited from class java.lang.Object

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

Methods inherited from interface com.emphysic.myriad.core.data.util.ObjectMap

init, initPreviousVersion, initUnknownVersion, load, read, save, write

Constructor Detail

FeatureScalingOperation

public FeatureScalingOperation()

Creates a new FeatureScalingOperation.

FeatureScalingOperation

public FeatureScalingOperation(java.util.List<? extends Dataset> datasets)

Creates a new scaler and fits with the provided data.

Parameters:

datasets - data with which to fit the scaler

Method Detail

run

public Dataset run(Dataset input)

Scales the input data by substracting the feature means and dividing by their standard deviations.

Specified by:

run in interface DatasetOperation

Parameters:

input - Dataset to scale

Returns:

scaled Dataset

Throws:

java.lang.RuntimeException - if the scaler hasn't been trained yet

See Also:

fit(java.util.List<? extends com.emphysic.myriad.core.data.io.Dataset>)

fit

public void fit(java.util.List<? extends Dataset> datasets)

"Trains" (i.e. calculates feature means and standard deviations) from the specified data. Used when the entire sample set is available up front as it replaces the current values.

Parameters:

datasets - data to use

partial_fit

public void partial_fit(java.util.List<? extends Dataset> datasets)

Incremental training - keeps approximations of feature means and standard deviations. Used when not all of the samples are available up front.

Parameters:

datasets - new batch of samples of size >=2 (approximations require 2 or more elements)

getFeatureMeans

public java.util.List<java.lang.Double> getFeatureMeans()

Returns the feature means

Returns:

feature means

setFeatureMeans

public void setFeatureMeans(java.util.List<java.lang.Double> featureMeans)

Sets the feature means

Parameters:

featureMeans - means of the features

getFeatureStds

public java.util.List<java.lang.Double> getFeatureStds()

Returns the feature standard deviations

Returns:

feature standard deviations

setFeatureStds

public void setFeatureStds(java.util.List<java.lang.Double> featureStds)

Sets the feature standard deviations

Parameters:

featureStds - standard deviations of the features

getSerializationVersion

public long getSerializationVersion()

Description copied from interface: ObjectMap

Returns the current version of the serialization format. Increments when the serialization format changes, i.e. backwards compatibility is broken.

Specified by:

getSerializationVersion in interface ObjectMap

Returns:

serialization version

getVersion

public int getVersion()

Description copied from interface: ObjectMap

Returns the current class version. Class version increments when the class changes but doesn't necessarily break compatibility with serialization.

Specified by:

getVersion in interface ObjectMap

Returns:

class version

getObjectMap

public java.util.Map<java.lang.String,java.lang.Object> getObjectMap()

Description copied from interface: ObjectMap

Creates a map of the important fields for the instance, suitable for serialization.

Specified by:

getObjectMap in interface ObjectMap

Returns:

"object map" of the instance

initCurrentVersion

public void initCurrentVersion(java.util.Map<java.lang.String,java.lang.Object> objectMap)

Description copied from interface: ObjectMap

Initializes an instance with a current-version object graph.

Specified by:

initCurrentVersion in interface ObjectMap

Parameters:

objectMap - object graph for initialization

getOnlineStats

public OnlineStats getOnlineStats()

Returns the online statistics approximator.

Returns:

OnlineStats instance

main

public static void main(java.lang.String[] args)