Package | Description |
---|---|
gov.sandia.cognition.learning.algorithm.ensemble |
Provides ensemble methods.
|
gov.sandia.cognition.learning.algorithm.perceptron |
Provides the Perceptron algorithm and some of its variations.
|
gov.sandia.cognition.learning.algorithm.perceptron.kernel | |
gov.sandia.cognition.learning.function.categorization |
Provides functions that output a discrete set of categories.
|
Modifier and Type | Class and Description |
---|---|
class |
WeightedBinaryEnsemble<InputType,MemberType extends Evaluator<? super InputType,? extends java.lang.Boolean>>
The
WeightedBinaryEnsemble class implements an Ensemble of
BinaryCategorizer objects where each categorizer is assigned a
weight and the category is selected by choosing the one with the largest
sum of weights. |
Modifier and Type | Field and Description |
---|---|
protected java.util.Collection<BinaryCategorizer<? super InputType>> |
BinaryCategorizerSelector.categorizers
The collection of categorizers to evaluate and select from.
|
Modifier and Type | Method and Description |
---|---|
BinaryCategorizer<? super InputType> |
BinaryCategorizerSelector.learn(java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Boolean>> data)
Selects the BinaryCategorizer from its list of categorizers that
minimizes the weighted error on the given set of weighted input-output
pairs.
|
Modifier and Type | Method and Description |
---|---|
java.util.Collection<BinaryCategorizer<? super InputType>> |
BinaryCategorizerSelector.getCategorizers()
Gets the collection of categorizers that the learner selects from.
|
Modifier and Type | Method and Description |
---|---|
void |
BinaryCategorizerSelector.setCategorizers(java.util.Collection<BinaryCategorizer<? super InputType>> categorizers)
Gets the collection of categorizers that the learner selects from.
|
Constructor and Description |
---|
BinaryCategorizerSelector(java.util.Collection<BinaryCategorizer<? super InputType>> categorizers)
Creates a new instance of
BinaryCategorizerSelector . |
Modifier and Type | Class and Description |
---|---|
static class |
OnlineShiftingPerceptron.LinearResult
This is the result learned by the shifting perceptron.
|
Modifier and Type | Class and Description |
---|---|
static class |
Forgetron.Result<InputType>
The result object learned by the
Forgetron , which extends
the DefaultKernelBinaryCategorizer with some additional state
information needed in the update step. |
Modifier and Type | Interface and Description |
---|---|
interface |
ConfidenceWeightedBinaryCategorizer
Interface for a confidence-weighted binary categorizer, which defines a
distribution over linear binary categorizers.
|
interface |
DiscriminantBinaryCategorizer<InputType>
Interface for a linear discriminant categorizer in the binary categorization
domain.
|
interface |
ThresholdBinaryCategorizer<InputType>
Interface for a binary categorizer that uses a threshold to determine the
categorization.
|
Modifier and Type | Class and Description |
---|---|
class |
AbstractBinaryCategorizer<InputType>
The
AbstractBinaryCategorizer implements the commonality of
the BinaryCategorizer , holding the collection of possible
values. |
class |
AbstractConfidenceWeightedBinaryCategorizer
Unit tests for class AbstractConfidenceWeightedBinaryCategorizer.
|
class |
AbstractDiscriminantBinaryCategorizer<InputType>
An abstract implementation of the
DiscriminantBinaryCategorizer
interface. |
class |
AbstractThresholdBinaryCategorizer<InputType>
Categorizer that first maps the input space onto a real value, then
uses a threshold to map the result onto lowValue (for strictly less than the
threshold) or highValue (for greater than or equal to the threshold).
|
class |
DefaultConfidenceWeightedBinaryCategorizer
A default implementation of the
ConfidenceWeightedBinaryCategorizer
that stores a full mean and covariance matrix. |
class |
DefaultKernelBinaryCategorizer<InputType>
A default implementation of the
KernelBinaryCategorizer that uses
the standard way of representing the examples (supports) using a
DefaultWeightedValue . |
class |
DiagonalConfidenceWeightedBinaryCategorizer
A confidence-weighted linear predictor with a diagonal covariance,
which is stored as a vector.
|
class |
FisherLinearDiscriminantBinaryCategorizer
A Fisher Linear Discriminant classifier, which creates an optimal linear
separating plane between two Gaussian classes of different covariances.
|
class |
KernelBinaryCategorizer<InputType,EntryType extends WeightedValue<? extends InputType>>
The
KernelBinaryCategorizer class implements a binary
categorizer that uses a kernel to do its categorization. |
class |
LinearBinaryCategorizer
The
LinearBinaryCategorizer class implements a binary
categorizer that is implemented by a linear function. |
class |
ScalarFunctionToBinaryCategorizerAdapter<InputType>
Adapts a scalar function to be a categorizer using a threshold.
|
class |
ScalarThresholdBinaryCategorizer
The
ScalarThresholdBinaryCategorizer class implements a binary
categorizer that uses a threshold to categorize a given double. |
class |
VectorElementThresholdCategorizer
The
VectorElementThresholdCategorizer class implements a
BinaryCategorizer that categorizes an input vector by applying a
threshold to an element in a the vector. |