类 RegOptimizer
java.lang.Object
weka.classifiers.functions.supportVector.RegOptimizer
- 所有已实现的接口:
Serializable
,OptionHandler
,RevisionHandler
- 直接已知子类:
RegSMO
Base class implementation for learning algorithm of SMOreg
Valid options are:
-L <double> The epsilon parameter in epsilon-insensitive loss function. (default 1.0e-3)
-W <double> The random number seed. (default 1)
- 版本:
- $Revision: 11614 $
- 作者:
- Remco Bouckaert (remco@cs.waikato.ac.nz,rrb@xm.co.nz)
- 另请参阅:
-
字段概要
字段修饰符和类型字段说明double[]
alpha and alpha* arrays containing weights for solving dual problemdouble[]
-
构造器概要
构造器 -
方法概要
修饰符和类型方法说明void
buildClassifier
(Instances data) learn SVM parameters from data.Returns the tip text for this propertyint
return the number of kernel cache hitsdouble
Get the value of epsilon parameter of the epsilon insensitive loss function.long
returns the number of kernel evaluationsString[]
Gets the current settings of the classifier.Returns the revision string.int
getSeed()
Gets the current seed value for the random number generatorGets an enumeration describing the available options.boolean
flag to indicate whether the model was built yetReturns the tip text for this propertyvoid
setEpsilonParameter
(double v) Set the value of epsilon parameter of the epsilon insensitive loss function.void
setOptions
(String[] options) Parses a given list of options.void
setSeed
(int value) Sets the seed value for the random number generatorvoid
sets the parent SVMdouble
toString()
Prints out the classifier.
-
字段详细资料
-
m_alpha
public double[] m_alphaalpha and alpha* arrays containing weights for solving dual problem -
m_alphaStar
public double[] m_alphaStar
-
-
构造器详细资料
-
RegOptimizer
public RegOptimizer()the default constructor
-
-
方法详细资料
-
listOptions
Gets an enumeration describing the available options.- 指定者:
listOptions
在接口中OptionHandler
- 返回:
- an enumeration of all the available options.
-
setOptions
Parses a given list of options. Valid options are:-L <double> The epsilon parameter in epsilon-insensitive loss function. (default 1.0e-3)
-W <double> The random number seed. (default 1)
- 指定者:
setOptions
在接口中OptionHandler
- 参数:
options
- the list of options as an array of strings- 抛出:
Exception
- if an option is not supported
-
getOptions
Gets the current settings of the classifier.- 指定者:
getOptions
在接口中OptionHandler
- 返回:
- an array of strings suitable for passing to setOptions
-
modelBuilt
public boolean modelBuilt()flag to indicate whether the model was built yet- 返回:
- true if the model was built
-
setSMOReg
sets the parent SVM- 参数:
value
- the parent SVM
-
getKernelEvaluations
public long getKernelEvaluations()returns the number of kernel evaluations- 返回:
- the number of kernel evaluations
-
getCacheHits
public int getCacheHits()return the number of kernel cache hits- 返回:
- the number of hits
-
buildClassifier
learn SVM parameters from data. Subclasses should implement something more interesting.- 参数:
data
- the data to work with- 抛出:
Exception
- always an Exceoption since subclasses must override it
-
SVMOutput
- 参数:
inst
-- 返回:
- 抛出:
Exception
-
seedTipText
Returns the tip text for this property- 返回:
- tip text for this property suitable for displaying in the explorer/experimenter gui
-
getSeed
public int getSeed()Gets the current seed value for the random number generator- 返回:
- the seed value
-
setSeed
public void setSeed(int value) Sets the seed value for the random number generator- 参数:
value
- the seed value
-
epsilonParameterTipText
Returns the tip text for this property- 返回:
- tip text for this property suitable for displaying in the explorer/experimenter gui
-
getEpsilonParameter
public double getEpsilonParameter()Get the value of epsilon parameter of the epsilon insensitive loss function.- 返回:
- Value of epsilon parameter.
-
setEpsilonParameter
public void setEpsilonParameter(double v) Set the value of epsilon parameter of the epsilon insensitive loss function.- 参数:
v
- Value to assign to epsilon parameter.
-
toString
Prints out the classifier. -
getRevision
Returns the revision string.- 指定者:
getRevision
在接口中RevisionHandler
- 返回:
- the revision
-