Package gov.sandia.cognition.learning.algorithm.perceptron.kernel

Interface Summary
BudgetedKernelBinaryCategorizerLearner Interface for a budgeted kernel binary categorizer learner.
 

Class Summary
AbstractOnlineBudgetedKernelBinaryCategorizerLearner<InputType> An abstract implementation of the BudgetedKernelBinaryCategorizerLearner for online learners.
AbstractOnlineKernelBinaryCategorizerLearner<InputType> An abstract class for an online kernel binary categorizer learner.
Forgetron<InputType> An implementation of the "self-tuned" Forgetron algorithm, which is an online budgeted kernel binary categorizer learner.
Forgetron.Basic<InputType> An implementation of the "basic" Forgetron algorithm, which is an online budgeted kernel binary categorizer learner.
Forgetron.Greedy<InputType> An implementation of the "greedy" Forgetron algorithm, which is an online budgeted kernel binary categorizer learner.
Forgetron.Result<InputType> The result object learned by the Forgetron, which extends the DefaultKernelBinaryCategorizer with some additional state information needed in the update step.
KernelAdatron<InputType> The KernelAdatron class implements an online version of the Support Vector Machine learning algorithm.
KernelBinaryCategorizerOnlineLearnerAdapter<InputType> A wrapper class for a KernelizableBinaryCategorizerOnlineLearner that allows it to be used as a batch or incremental learner over the input type directly, rather than using utility methods.
KernelPerceptron<InputType> The KernelPerceptron class implements the kernel version of the Perceptron algorithm.
OnlineKernelPerceptron<InputType> An implementation of the online version of the Perceptron algorithm.
OnlineKernelRandomizedBudgetPerceptron<InputType> An implementation of a fixed-memory kernel Perceptron algorithm.
Projectron<InputType> An implementation of the Projectron algorithm, which is an online kernel binary categorizer learner that has a budget parameter tuned by the eta parameter.
Projectron.LinearSoftMargin<InputType> An implementation of the Projectron++ algorithm, which is an online kernel binary categorizer learner that has a budget parameter tuned by the eta parameter.
RemoveOldestKernelPerceptron<InputType> A budget kernel Perceptron that always removes the oldest item.
Stoptron<InputType> An online, budgeted, kernel version of the Perceptron algorithm that stops learning once it has reached its budget.