gov.sandia.cognition.learning.algorithm.perceptron

Interface KernelizableBinaryCategorizerOnlineLearner

All Superinterfaces:

BatchAndIncrementalLearner<InputOutputPair<? extends Vectorizable,java.lang.Boolean>,LinearBinaryCategorizer>, BatchLearner<java.util.Collection<? extends InputOutputPair<? extends Vectorizable,java.lang.Boolean>>,LinearBinaryCategorizer>, java.lang.Cloneable, CloneableSerializable, IncrementalLearner<InputOutputPair<? extends Vectorizable,java.lang.Boolean>,LinearBinaryCategorizer>, java.io.Serializable, SupervisedBatchAndIncrementalLearner<Vectorizable,java.lang.Boolean,LinearBinaryCategorizer>, SupervisedBatchLearner<Vectorizable,java.lang.Boolean,LinearBinaryCategorizer>, SupervisedIncrementalLearner<Vectorizable,java.lang.Boolean,LinearBinaryCategorizer>

All Known Implementing Classes:

AbstractKernelizableBinaryCategorizerOnlineLearner, AbstractLinearCombinationOnlineLearner, AggressiveRelaxedOnlineMaximumMarginAlgorithm, Ballseptron, OnlineBinaryMarginInfusedRelaxedAlgorithm, OnlinePassiveAggressivePerceptron, OnlinePassiveAggressivePerceptron.AbstractSoftMargin, OnlinePassiveAggressivePerceptron.LinearSoftMargin, OnlinePassiveAggressivePerceptron.QuadraticSoftMargin, OnlinePerceptron, OnlineRampPassiveAggressivePerceptron, OnlineShiftingPerceptron, RelaxedOnlineMaximumMarginAlgorithm

public interface KernelizableBinaryCategorizerOnlineLearner
extends SupervisedBatchAndIncrementalLearner<Vectorizable,java.lang.Boolean,LinearBinaryCategorizer>

Interface for an online learner of a linear binary categorizer that can also be used with a kernel function. Thus, there are a second set of methods that are similar to the ones for the normal learner that are specifically for kernel learning. The companion to this class is LinearizableBinaryCategorizerOnlineLearner.

Since:

3.3.0

Author:

Justin Basilico

See Also:

LinearizableBinaryCategorizerOnlineLearner

Method Summary

Modifier and Type	Method and Description
<InputType> DefaultKernelBinaryCategorizer<InputType>	createInitialLearnedObject(Kernel<? super InputType> kernel) Creates the initial learned object with a given kernel.
<InputType> SupervisedBatchAndIncrementalLearner<InputType,java.lang.Boolean,DefaultKernelBinaryCategorizer<InputType>>	createKernelLearner(Kernel<? super InputType> kernel) Creates a new kernel-based learner using the standard learning interfaces based on this learner and its parameters.
<InputType> DefaultKernelBinaryCategorizer<InputType>	learn(Kernel<? super InputType> kernel, java.lang.Iterable<? extends InputOutputPair<? extends InputType,java.lang.Boolean>> data) Run this algorithm on a batch of data using the given kernel function.
<InputType> void	update(DefaultKernelBinaryCategorizer<InputType> target, InputOutputPair<? extends InputType,java.lang.Boolean> data) Performs a kernel-based incremental update step on the given object using the given supervised data.
<InputType> void	update(DefaultKernelBinaryCategorizer<InputType> target, InputType input, boolean output) Performs a kernel-based incremental update step on the given object using the given supervised data.
<InputType> void	update(DefaultKernelBinaryCategorizer<InputType> target, InputType input, java.lang.Boolean output) Performs a kernel-based incremental update step on the given object using the given supervised data.
<InputType> void	update(DefaultKernelBinaryCategorizer<InputType> target, java.lang.Iterable<? extends InputOutputPair<? extends InputType,java.lang.Boolean>> data) Performs a kernel-based incremental update step on the given object using the given supervised data.

Methods inherited from interface gov.sandia.cognition.learning.algorithm.SupervisedIncrementalLearner

update

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchAndIncrementalLearner

learn

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchLearner

learn

Methods inherited from interface gov.sandia.cognition.learning.algorithm.IncrementalLearner

createInitialLearnedObject, update, update

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Method Detail

createInitialLearnedObject

<InputType> DefaultKernelBinaryCategorizer<InputType> createInitialLearnedObject(Kernel<? super InputType> kernel)

Creates the initial learned object with a given kernel.

Type Parameters:

InputType - The input type for supervised learning. Will be passed to the kernel function.

Parameters:

kernel - The kernel function to use.

Returns:

A new, empty learned object.

update

<InputType> void update(DefaultKernelBinaryCategorizer<InputType> target,
                        java.lang.Iterable<? extends InputOutputPair<? extends InputType,java.lang.Boolean>> data)

Performs a kernel-based incremental update step on the given object using the given supervised data.

Type Parameters:

InputType - The input type for supervised learning. Will be passed to the kernel function.

Parameters:

target - The target object to update.

data - The supervised training data.

update

<InputType> void update(DefaultKernelBinaryCategorizer<InputType> target,
                        InputOutputPair<? extends InputType,java.lang.Boolean> data)

Performs a kernel-based incremental update step on the given object using the given supervised data.

Type Parameters:

InputType - The input type for supervised learning. Will be passed to the kernel function.

Parameters:

target - The target object to update.

data - The supervised training data.

update

<InputType> void update(DefaultKernelBinaryCategorizer<InputType> target,
                        InputType input,
                        java.lang.Boolean output)

Performs a kernel-based incremental update step on the given object using the given supervised data.

Type Parameters:

InputType - The input type for supervised learning. Will be passed to the kernel function.

Parameters:

target - The target object to update.

input - The supervised input value.

output - The supervised output value (label).

update

<InputType> void update(DefaultKernelBinaryCategorizer<InputType> target,
                        InputType input,
                        boolean output)

Performs a kernel-based incremental update step on the given object using the given supervised data.

Type Parameters:

InputType - The input type for supervised learning. Will be passed to the kernel function.

Parameters:

target - The target object to update.

input - The supervised input value.

output - The supervised output value (label).

learn

<InputType> DefaultKernelBinaryCategorizer<InputType> learn(Kernel<? super InputType> kernel,
                                                            java.lang.Iterable<? extends InputOutputPair<? extends InputType,java.lang.Boolean>> data)

Run this algorithm on a batch of data using the given kernel function.

Type Parameters:

InputType - The input type for supervised learning. Will be passed to the kernel function.

Parameters:

kernel - The kernel function to use.

data - The supervised training data.

Returns:

A new object trained on the given data.

createKernelLearner

<InputType> SupervisedBatchAndIncrementalLearner<InputType,java.lang.Boolean,DefaultKernelBinaryCategorizer<InputType>> createKernelLearner(Kernel<? super InputType> kernel)

Creates a new kernel-based learner using the standard learning interfaces based on this learner and its parameters.

Type Parameters:

InputType - The input type for supervised learning. Will be passed to the kernel function.

Parameters:

kernel - The kernel function to use.

Returns:

A kernel-based version of this learning algorithm.