smile.regression

Class SVR<T>

java.lang.Object

smile.regression.SVR<T>

All Implemented Interfaces:

Regression<T>

public class SVR<T>
extends Object
implements Regression<T>

Support vector regression. Like SVMs for classification, the model produced by SVR depends only on a subset of the training data, because the cost function ignores any training data close to the model prediction (within a threshold ε).

References

1. A. J Smola and B. Scholkopf. A Tutorial on Support Vector Regression.
2. Gary William Flake and Steve Lawrence. Efficient SVM Regression Training with SMO.
3. Christopher J. C. Burges. A Tutorial on Support Vector Machines for Pattern Recognition. Data Mining and Knowledge Discovery 2:121-167, 1998.
4. John Platt. Sequential Minimal Optimization: A Fast Algorithm for Training Support Vector Machines.
5. Rong-En Fan, Pai-Hsuen, and Chih-Jen Lin. Working Set Selection Using Second Order Information for Training Support Vector Machines. JMLR, 6:1889-1918, 2005.
6. Antoine Bordes, Seyda Ertekin, Jason Weston and Leon Bottou. Fast Kernel Classifiers with Online and Active Learning, Journal of Machine Learning Research, 6:1579-1619, 2005.
7. Tobias Glasmachers and Christian Igel. Second Order SMO Improves SVM Online and Active Learning.
8. Chih-Chung Chang and Chih-Jen Lin. LIBSVM: a Library for Support Vector Machines.

Author:

Haifeng Li

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	SVR.Trainer<T> Trainer for support vector regression.

Constructor Summary

Constructors

Constructor and Description
SVR(T[] x, double[] y, double[] weight, MercerKernel<T> kernel, double eps, double C) Constructor.
SVR(T[] x, double[] y, double[] weight, MercerKernel<T> kernel, double eps, double C, double tol) Constructor.
SVR(T[] x, double[] y, MercerKernel<T> kernel, double eps, double C) Constructor.
SVR(T[] x, double[] y, MercerKernel<T> kernel, double eps, double C, double tol) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
double	getC() Returns the soft margin penalty parameter.
double	getEpsilon() Returns the loss function error threshold.
double	getTolerance() Returns the tolerance of convergence test.
double	predict(T x) Predicts the dependent variable of an instance.

Methods inherited from class java.lang.Object

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

Constructor Detail

SVR

public SVR(T[] x,
   double[] y,
   MercerKernel<T> kernel,
   double eps,
   double C)

Constructor.

Parameters:

x - training instances.

y - response variable.

kernel - the kernel function.

eps - the loss function error threshold.

C - the soft margin penalty parameter.

SVR

public SVR(T[] x,
   double[] y,
   double[] weight,
   MercerKernel<T> kernel,
   double eps,
   double C)

Constructor.

Parameters:

x - training instances.

y - response variable.

weight - positive instance weight. The soft margin penalty parameter for instance i will be weight[i] * C.

kernel - the kernel function.

eps - the loss function error threshold.

C - the soft margin penalty parameter.

SVR

public SVR(T[] x,
   double[] y,
   MercerKernel<T> kernel,
   double eps,
   double C,
   double tol)

Constructor.

Parameters:

x - training instances.

y - response variable.

kernel - the kernel function.

eps - the loss function error threshold.

C - the soft margin penalty parameter.

tol - the tolerance of convergence test.

SVR

public SVR(T[] x,
   double[] y,
   double[] weight,
   MercerKernel<T> kernel,
   double eps,
   double C,
   double tol)

Constructor.

Parameters:

x - training instances.

y - response variable.

weight - positive instance weight. The soft margin penalty parameter for instance i will be weight[i] * C.

kernel - the kernel function.

eps - the loss function error threshold.

C - the soft margin penalty parameter.

tol - the tolerance of convergence test.

Method Detail

predict

public double predict(T x)

Description copied from interface: Regression

Predicts the dependent variable of an instance.

Specified by:

predict in interface Regression<T>

Parameters:

x - the instance.

Returns:

the predicted value of dependent variable.

getC

public double getC()

Returns the soft margin penalty parameter.

Returns:

the soft margin penalty parameter.

getEpsilon

public double getEpsilon()

Returns the loss function error threshold.

Returns:

the loss function error threshold.

getTolerance

public double getTolerance()

Returns the tolerance of convergence test.

Returns:

tolerance of convergence test.