A B C E F G J K L M N O P R S T V W X

A

alphafile - Variable in class jnisvmlight.LearnParam

File to store optimal alphas in.

argc - Variable in class jnisvmlight.LearnParam

The cardinality of the command line parameters.

argv - Variable in class jnisvmlight.LearnParam

Optionally simulates a simple command shell-like usage and transfers the command line parameters to SVM-light.

B

biased_hyperplane - Variable in class jnisvmlight.LearnParam

If nonzero, use hyperplane w*x+b=0 otherwise w*x=0.

C

CLASSIFICATION - Static variable in class jnisvmlight.LearnParam

Trains a classification model.

classify(FeatureVector) - Method in class jnisvmlight.SVMLightModel

classifyNative(FeatureVector) - Method in class jnisvmlight.SVMLightInterface

Performs a classifcation step as a native call to SVM-light.

coef_const - Variable in class jnisvmlight.KernelParam

Constant coefficient for extended kernels.

coef_lin - Variable in class jnisvmlight.KernelParam

Linear coefficient for extended kernels.

compute_loo - Variable in class jnisvmlight.LearnParam

If nonzero, computes leave-one-outestimates.

custom - Variable in class jnisvmlight.KernelParam

For user supplied kernel.

E

eps - Variable in class jnisvmlight.LearnParam

Regression epsilon (eps=1.0 for classification).

epsilon_a - Variable in class jnisvmlight.LearnParam

Tolerable error on alphas at bounds.

epsilon_const - Variable in class jnisvmlight.LearnParam

Tolerable error on eq-constraint.

epsilon_crit - Variable in class jnisvmlight.LearnParam

Tolerable error for distances used in stopping criterion.

epsilon_shrink - Variable in class jnisvmlight.LearnParam

How much a multiplier should be above zero for shrinking.

evaluate(FeatureVector, FeatureVector) - Method in class jnisvmlight.Kernel

evaluate(FeatureVector, FeatureVector) - Method in class jnisvmlight.LinearKernel

evaluate(FeatureVector, FeatureVector) - Method in class jnisvmlight.PolynomialKernel

evaluate(FeatureVector, FeatureVector) - Method in class jnisvmlight.RadialBaseKernel

evaluate(FeatureVector, FeatureVector) - Method in class jnisvmlight.SigmoidKernel

ExtendedKernel - Class in jnisvmlight

Abstract class for an extended kernel.

ExtendedKernel() - Constructor for class jnisvmlight.ExtendedKernel

ExtendedKernel(Kernel, double, double) - Constructor for class jnisvmlight.ExtendedKernel

F

FeatureVector - Class in jnisvmlight

A feature vector.

FeatureVector(double, int[], double[]) - Constructor for class jnisvmlight.FeatureVector

FeatureVector(int) - Constructor for class jnisvmlight.FeatureVector

FeatureVector(int[], double[]) - Constructor for class jnisvmlight.FeatureVector

G

getConstant() - Method in class jnisvmlight.ExtendedKernel

getCosine(FeatureVector) - Method in class jnisvmlight.FeatureVector

Returns the cosine similarity between two feature vectors.

getDimAt(int) - Method in class jnisvmlight.FeatureVector

getFactor() - Method in class jnisvmlight.FeatureVector

getKernelParameters() - Method in class jnisvmlight.TrainingParameters

getL1Norm() - Method in class jnisvmlight.FeatureVector

Returns the linear norm factor of this vector's values (i.e., the sum of it's values).

getL2Norm() - Method in class jnisvmlight.FeatureVector

Returns the L2 norm factor of this vector's values.

getLabel() - Method in class jnisvmlight.LabeledFeatureVector

getLabeledFeatureVectorsFromURL(URL, int) - Static method in class jnisvmlight.SVMLightInterface

Reads a set of labeled training vectors from a URL.

getLearningParameters() - Method in class jnisvmlight.TrainingParameters

getLinearWeights() - Method in class jnisvmlight.SVMLightModel

Returns a vector with the weights of all features for linear kernels.

getMultiplier() - Method in class jnisvmlight.ExtendedKernel

getNestedKernel() - Method in class jnisvmlight.Kernel

getOrder() - Method in class jnisvmlight.PolynomialKernel

getTrainingParameters() - Method in class jnisvmlight.SVMLightInterface

getValueAt(int) - Method in class jnisvmlight.FeatureVector

getWidth() - Method in class jnisvmlight.RadialBaseKernel

J

jnisvmlight - package jnisvmlight

K

Kernel - Class in jnisvmlight

Abstract kernel class.

Kernel() - Constructor for class jnisvmlight.Kernel

Kernel(Kernel) - Constructor for class jnisvmlight.Kernel

kernel_cache_size - Variable in class jnisvmlight.LearnParam

Size of kernel cache in megabytes.

kernel_type - Variable in class jnisvmlight.KernelParam

Selects between LINEAR, POLYNOMIAL, RBF, or SIGMOID kernel type.

KernelParam - Class in jnisvmlight

Kernel parameters used by SVM-light.

KernelParam() - Constructor for class jnisvmlight.KernelParam

Initializes the kernel parameters with the default SVM-light values.

L

LabeledFeatureVector - Class in jnisvmlight

A labeled feature vector.

LabeledFeatureVector() - Constructor for class jnisvmlight.LabeledFeatureVector

LabeledFeatureVector(double, int) - Constructor for class jnisvmlight.LabeledFeatureVector

LabeledFeatureVector(double, int[], double[]) - Constructor for class jnisvmlight.LabeledFeatureVector

LearnParam - Class in jnisvmlight

Learning parameters as denoted by SVM-light.

LearnParam() - Constructor for class jnisvmlight.LearnParam

Initializes the learning parameters with the default SVM-light values.

LINEAR - Static variable in class jnisvmlight.KernelParam

Uses a linear kernel type.

LinearKernel - Class in jnisvmlight

A linear kernel.

LinearKernel() - Constructor for class jnisvmlight.LinearKernel

M

m_a - Variable in class jnisvmlight.ExtendedKernel

m_c - Variable in class jnisvmlight.ExtendedKernel

m_dims - Variable in class jnisvmlight.FeatureVector

m_factor - Variable in class jnisvmlight.FeatureVector

m_kernel - Variable in class jnisvmlight.Kernel

m_label - Variable in class jnisvmlight.LabeledFeatureVector

m_linWeights - Variable in class jnisvmlight.SVMLightModel

m_tp - Variable in class jnisvmlight.SVMLightInterface

m_vals - Variable in class jnisvmlight.FeatureVector

maxiter - Variable in class jnisvmlight.LearnParam

Number of iterations after which the optimizer terminates, if there was no progress in maxdiff.

N

normalizeL1() - Method in class jnisvmlight.FeatureVector

Performs a linear normalization to the value 1.

normalizeL1(double) - Method in class jnisvmlight.FeatureVector

Performs a linear normalization to the given norm value.

normalizeL2() - Method in class jnisvmlight.FeatureVector

Performs an L2 normalization to the value 1.

O

opt_precision - Variable in class jnisvmlight.LearnParam

Precision of solver, set to e.g.

OPTIMIZATION - Static variable in class jnisvmlight.LearnParam

Trains on general set of constraints.

P

poly_degree - Variable in class jnisvmlight.KernelParam

Degree of polynomial kernel.

POLYNOMIAL - Static variable in class jnisvmlight.KernelParam

Uses a polynomial kernel type.

PolynomialKernel - Class in jnisvmlight

A polynomial kernel.

PolynomialKernel() - Constructor for class jnisvmlight.PolynomialKernel

PolynomialKernel(Kernel, double, double, double) - Constructor for class jnisvmlight.PolynomialKernel

predfile - Variable in class jnisvmlight.LearnParam

File for predictions on unlabeled examples in transduction.

R

RadialBaseKernel - Class in jnisvmlight

A radial base kernel.

RadialBaseKernel() - Constructor for class jnisvmlight.RadialBaseKernel

RadialBaseKernel(Kernel, double) - Constructor for class jnisvmlight.RadialBaseKernel

RANKING - Static variable in class jnisvmlight.LearnParam

Trains a ranking model.

RBF - Static variable in class jnisvmlight.KernelParam

Use a radial base kernel type.

rbf_gamma - Variable in class jnisvmlight.KernelParam

Gamma constant for a radial base kernel.

readSVMLightModelFromURL(URL) - Static method in class jnisvmlight.SVMLightModel

Reads an SVM-light model from a URL and creates an SVMLightModel object in Java.

REGRESSION - Static variable in class jnisvmlight.LearnParam

Trains a regression model.

remove_inconsistent - Variable in class jnisvmlight.LearnParam

Exclude examples with alpha at C and retrain.

rho - Variable in class jnisvmlight.LearnParam

Parameter in xi/alpha-estimates and for pruning leave-one-out range [1..2].

S

setConstant(double) - Method in class jnisvmlight.ExtendedKernel

setFactor(double) - Method in class jnisvmlight.FeatureVector

setFeatures(int[], double[]) - Method in class jnisvmlight.FeatureVector

setFeatures(double, int[], double[]) - Method in class jnisvmlight.LabeledFeatureVector

setKernelParameters(KernelParam) - Method in class jnisvmlight.TrainingParameters

setLabel(double) - Method in class jnisvmlight.LabeledFeatureVector

setLearningParameters(LearnParam) - Method in class jnisvmlight.TrainingParameters

setMultiplier(double) - Method in class jnisvmlight.ExtendedKernel

setNestedKernel(Kernel) - Method in class jnisvmlight.Kernel

setOrder(double) - Method in class jnisvmlight.PolynomialKernel

setThreshold(double) - Method in class jnisvmlight.SVMLightModel

setWidth(double) - Method in class jnisvmlight.RadialBaseKernel

sharedslack - Variable in class jnisvmlight.LearnParam

If nonzero, it will use the shared slack variable mode.

SIGMOID - Static variable in class jnisvmlight.KernelParam

Uses as sigmoid kernel type.

SigmoidKernel - Class in jnisvmlight

A kernel with sigmoid smoothing.

SigmoidKernel() - Constructor for class jnisvmlight.SigmoidKernel

SigmoidKernel(Kernel, double, double) - Constructor for class jnisvmlight.SigmoidKernel

size() - Method in class jnisvmlight.FeatureVector

skip_final_opt_check - Variable in class jnisvmlight.LearnParam

Do not check KT-Conditions at the end of optimization for examples removed by shrinking.

SORT_INPUT_VECTORS - Static variable in class jnisvmlight.SVMLightInterface

Apply an in-place quicksort prior to each native training call to SVM-light.

svm_c - Variable in class jnisvmlight.LearnParam

Upper bound C on alphas.

svm_c_factor - Variable in class jnisvmlight.LearnParam

Increase C by this factor every step.

svm_c_steps - Variable in class jnisvmlight.LearnParam

Do so many steps for finding optimal C.

svm_cost - Variable in class jnisvmlight.LearnParam

Individual upper bounds for each var.

svm_costratio - Variable in class jnisvmlight.LearnParam

Factor to multiply C for positive examples.

svm_costratio_unlab - Variable in class jnisvmlight.LearnParam

svm_iter_to_shrink - Variable in class jnisvmlight.LearnParam

Iterations h after which an example can be removed by shrinking.

svm_maxqpsize - Variable in class jnisvmlight.LearnParam

Size q of working set.

svm_newvarsinqp - Variable in class jnisvmlight.LearnParam

New variables to enter the working set in each iteration.

svm_unlabbound - Variable in class jnisvmlight.LearnParam

SVMLightInterface - Class in jnisvmlight

The main interface class that transfers the training data to the SVM-light library by a native call.

SVMLightInterface() - Constructor for class jnisvmlight.SVMLightInterface

SVMLightModel - Class in jnisvmlight

SVM classifier model returned by SVM-light.

SVMLightModel(String, long, long, double, double, double, String, long, long, long, double, LabeledFeatureVector[]) - Constructor for class jnisvmlight.SVMLightModel

T

toString() - Method in class jnisvmlight.FeatureVector

toString() - Method in class jnisvmlight.LabeledFeatureVector

toString() - Method in class jnisvmlight.PolynomialKernel

toString() - Method in class jnisvmlight.RadialBaseKernel

toString() - Method in class jnisvmlight.SigmoidKernel

toString() - Method in class jnisvmlight.SVMLightModel

totwords - Variable in class jnisvmlight.LearnParam

Total amount of features.

TrainingParameters - Class in jnisvmlight

The TrainingParameters object combines the KernelParm and LearnParam objects.

TrainingParameters() - Constructor for class jnisvmlight.TrainingParameters

Initializes the training parameters with the default values for the kernel and the learning parameters.

TrainingParameters(LearnParam, KernelParam) - Constructor for class jnisvmlight.TrainingParameters

Initializes the training parameters with customized values for the kernel and the learning parameters.

TrainingParameters(String[]) - Constructor for class jnisvmlight.TrainingParameters

trainModel(LabeledFeatureVector[]) - Method in class jnisvmlight.SVMLightInterface

trainModel(LabeledFeatureVector[], String[]) - Method in class jnisvmlight.SVMLightInterface

trainModel(LabeledFeatureVector[], TrainingParameters) - Method in class jnisvmlight.SVMLightInterface

transduction_posratio - Variable in class jnisvmlight.LearnParam

Fraction of unlabeled examples to be classified as positives.

type - Variable in class jnisvmlight.LearnParam

Selects between CLASSIFICATION, REGRESSION, RANKING, or OPTIMIZATION mode.

V

verbosity - Variable in class jnisvmlight.LearnParam

The level of SVM-light debugging infos.

W

writeModelToFile(String) - Method in class jnisvmlight.SVMLightModel

Writes this SVMLightModel to a file.

X

xa_depth - Variable in class jnisvmlight.LearnParam

Parameter in xi/alpha-estimates upper bounding the number of SV the current alpha_t is distributed over.

A B C E F G J K L M N O P R S T V W X