Uses of Interface
com.indeed.vw.wrapper.api.parameters.SGDVowpalWabbitBuilder

Packages that use SGDVowpalWabbitBuilder

Package	Description
com.indeed.vw.wrapper.api	This package contains OOP api for wowpal wabbit jni wrapper.
com.indeed.vw.wrapper.api.parameters	API for creating vowpal wabbit parameters string.

Uses of SGDVowpalWabbitBuilder in com.indeed.vw.wrapper.api

Classes in com.indeed.vw.wrapper.api that implement SGDVowpalWabbitBuilder

Modifier and Type	Class and Description
static class	VowpalWabbit.Builder Builder for Vowpal Wabbit learner.

Methods in com.indeed.vw.wrapper.api that return SGDVowpalWabbitBuilder

Modifier and Type	Method and Description
static SGDVowpalWabbitBuilder	VowpalWabbit.builder() Create builder for Vowpal Wabbit learner

Uses of SGDVowpalWabbitBuilder in com.indeed.vw.wrapper.api.parameters

Methods in com.indeed.vw.wrapper.api.parameters that return SGDVowpalWabbitBuilder

Modifier and Type	Method and Description
SGDVowpalWabbitBuilder	UpdatesOptions.adaptive() use adaptive, individual learning rates.
SGDVowpalWabbitBuilder	OptionToExchangeRAMForQuality.bitPrecision(int bitsNum) number of bits in the feature table.
SGDVowpalWabbitBuilder	FeatureEngineeringFunctions.constant(double initialValue) Set initial value of constant
SGDVowpalWabbitBuilder	FeatureEngineeringFunctions.cubic(String firstNameSpace, String secondNamespace, String thirdNamespace) Create and use cubic features
SGDVowpalWabbitBuilder	FeatureSelectionOptions.featureMask(Path featureMask) Use existing regressor to determine which parameters may be updated.
SGDVowpalWabbitBuilder	OptionsToSaveAndLoadModel.finalRegressor(Path regressor) Final regressor
SGDVowpalWabbitBuilder	FeatureSelectionOptions.ftrl() FTRL: Follow the Proximal Regularized Leader
SGDVowpalWabbitBuilder	OptionsToSaveAndLoadModel.initialRegressor(Path initialRegressor) Initial regressor(s)
SGDVowpalWabbitBuilder	UpdatesOptions.invariant() use safe/importance aware updates.
SGDVowpalWabbitBuilder	RegularizationOptions.l1(double l1) l_1 lambda
SGDVowpalWabbitBuilder	RegularizationOptions.l2(double l2) l_2 lambda
SGDVowpalWabbitBuilder	UpdatesOptions.learningRate(double learningRate) Set learning rate
SGDVowpalWabbitBuilder	LinkAndLossOptions.link(Link link) Specify the link function: identity, logistic, glf1 or poisson (=identity)
SGDVowpalWabbitBuilder	LinkAndLossOptions.lossFunction(Loss loss) Specify the loss function to be used, uses squared by default.
SGDVowpalWabbitBuilder	FeatureEngineeringFunctions.lrqfa(String firstNamespace, String secondNamespace, int k) use low rank quadratic feature-aware weights
SGDVowpalWabbitBuilder	PredictionBoundaryOptions.maxPrediction(double max) Largest prediction to output
SGDVowpalWabbitBuilder	PredictionBoundaryOptions.minPrediction(double min) Smallest prediction to output
SGDVowpalWabbitBuilder	FeatureEngineeringFunctions.ngram(String namespace, int n) Generate N grams.
SGDVowpalWabbitBuilder	FeatureEngineeringFunctions.noconstant() Don't add a constant feature
SGDVowpalWabbitBuilder	UpdatesOptions.normalized() use per feature normalized updates
SGDVowpalWabbitBuilder	FeatureEngineeringFunctions.quadratic(String firstNameSpace, String secondNamespace) Create and use quadratic features
SGDVowpalWabbitBuilder	LinkAndLossOptions.quantileTau(double tau) Parameter \tau associated with Quantile loss.
SGDVowpalWabbitBuilder	MiscOptions.randomSeed(int seed) Seed random number generator
SGDVowpalWabbitBuilder	DebuggingOptions.readableModel(Path model) Output human-readable final regressor with numeric features
SGDVowpalWabbitBuilder	UpdatesOptions.sgd() use regular stochastic gradient descent update.
SGDVowpalWabbitBuilder	FeatureEngineeringFunctions.skips(String namespace, int n) Generate skips in N grams.
SGDVowpalWabbitBuilder	MiscOptions.testonly() Ignore label information and just test
SGDVowpalWabbitBuilder	DebuggingOptions.verbose() Make vowpal wabbit writing debug and performance information to stderr