com.workfusion.vds.sdk.ml.sklearn.linearsvc

Class LinearSupportVectorClassifier.Builder<B extends LinearSupportVectorClassifier.Builder,T extends LinearSupportVectorClassifier>

java.lang.Object

com.workfusion.vds.sdk.ml.python.BasePythonClassifier.Builder<LinearSupportVectorClassifier.Builder<B,T>,T>

com.workfusion.vds.sdk.ml.sklearn.linearsvc.LinearSupportVectorClassifier.Builder<B,T>

Direct Known Subclasses:

LinearSupportVectorParamGridClassifier.Builder

Enclosing class:

LinearSupportVectorClassifier

public static class LinearSupportVectorClassifier.Builder<B extends LinearSupportVectorClassifier.Builder,T extends LinearSupportVectorClassifier>
extends BasePythonClassifier.Builder<LinearSupportVectorClassifier.Builder<B,T>,T>

The type Builder.

Field Summary

Fields inherited from class com.workfusion.vds.sdk.ml.python.BasePythonClassifier.Builder

modelName, workingDir

Constructor Summary

Constructors

Constructor and Description
Builder() Instantiates a new Builder.

Method Summary

Modifier and Type	Method and Description
T	build() Build classifier.
B	classWeight(ClassWeightType classWeight) Sets class weight.
B	dual(boolean dual) Sets the algorithm to either solve the dual or primal optimization problem.
B	fitIntercept(boolean fitIntercept) Sets whether to calculate the intercept for this model.
B	interceptScaling(double interceptScaling) Sets the scaling coefficient of the synthetic features.
B	lossFunction(LossFunctionType lossFunction) Sets the loss function.
B	maxIterations(int maxIterations) Sets the maximum number of iterations to be run.
B	modelName(String name) Set a model name.
B	multiClass(MultiClassType multiClass) Sets the multi-class strategy if y contains more than two classes.
B	penalty(PenaltyType penalty) Sets the norm used in the penalization.
B	penaltyCoefficient(double penaltyCoefficient) Sets penalty coefficient of the error term.
B	seed(Integer seed) Sets the seed used by the random number generator.
B	tolerance(double tolerance) Sets the tolerance for stopping criteria.
B	verbose(boolean verbose) Sets whether verbose mode is used.

Methods inherited from class com.workfusion.vds.sdk.ml.python.BasePythonClassifier.Builder

createDefaultClassifierBridge, workingDir

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Builder

public Builder()

Instantiates a new Builder.

Method Detail

penalty

public B penalty(PenaltyType penalty)

Sets the norm used in the penalization.

Parameters:

penalty - the penalty (PenaltyType) (default: PenaltyType.L2)

Returns:

the builder

lossFunction

public B lossFunction(LossFunctionType lossFunction)

Sets the loss function.

Parameters:

lossFunction - the loss function (LossFunctionType) (default: LossFunctionType.SQUARED_HINGE)

Returns:

the builder

dual

public B dual(boolean dual)

Sets the algorithm to either solve the dual or primal optimization problem.

Parameters:

dual - the dual flag (default: true)

Returns:

the builder

tolerance

public B tolerance(double tolerance)

Sets the tolerance for stopping criteria.

Parameters:

tolerance - the tolerance (default: 1e-4)

Returns:

the builder

penaltyCoefficient

public B penaltyCoefficient(double penaltyCoefficient)

Sets penalty coefficient of the error term.

Parameters:

penaltyCoefficient - the penalty coefficient (default: 1.0)

Returns:

the builder

multiClass

public B multiClass(MultiClassType multiClass)

Sets the multi-class strategy if y contains more than two classes.

Parameters:

multiClass - the multi class strategy (MultiClassType) (default: MultiClassType.OVR)

Returns:

the builder

fitIntercept

public B fitIntercept(boolean fitIntercept)

Sets whether to calculate the intercept for this model. If set to false, no intercept will be used in calculations (i.e. data is expected to be already centered).

Parameters:

fitIntercept - the fit intercept flag (default: true)

Returns:

the builder

interceptScaling

public B interceptScaling(double interceptScaling)

Sets the scaling coefficient of the synthetic features.

Parameters:

interceptScaling - the scaling coefficient (default: 1.0)

Returns:

the builder

classWeight

public B classWeight(ClassWeightType classWeight)

Sets class weight. If null then all classes are supposed to have weight one.

Parameters:

classWeight - the class weight (ClassWeightType) (default: null)

Returns:

the builder

seed

public B seed(Integer seed)

Sets the seed used by the random number generator. If null then default random number generator is used (np.random).

Parameters:

seed - the seed (default: null)

Returns:

the builder

verbose

public B verbose(boolean verbose)

Sets whether verbose mode is used.

Parameters:

verbose - the verbose mode flag (default: false)

Returns:

the builder

maxIterations

public B maxIterations(int maxIterations)

Sets the maximum number of iterations to be run.

Parameters:

maxIterations - the number of iterations (default: 1000)

Returns:

the builder

modelName

public B modelName(String name)

Description copied from class: BasePythonClassifier.Builder

Set a model name.

Overrides:

modelName in class BasePythonClassifier.Builder<LinearSupportVectorClassifier.Builder<B extends LinearSupportVectorClassifier.Builder,T extends LinearSupportVectorClassifier>,T extends LinearSupportVectorClassifier>

Parameters:

name - the model name (String)

Returns:

the builder

build

public T build()

Description copied from class: BasePythonClassifier.Builder

Build classifier.

Specified by:

build in class BasePythonClassifier.Builder<LinearSupportVectorClassifier.Builder<B extends LinearSupportVectorClassifier.Builder,T extends LinearSupportVectorClassifier>,T extends LinearSupportVectorClassifier>

Returns:

the classifier instance