CDomainAdaptationSVM类参考

---

详细描述

class DomainAdaptiveSVM

在文件DomainAdaptationSVM.h第23行定义。

所有成员的列表。

公有成员
	CDomainAdaptationSVM ()
	CDomainAdaptationSVM (float64_t C, CKernel *k, CLabels *lab, CSVM *presvm, float64_t B)
virtual	~CDomainAdaptationSVM ()
void	init (CSVM *presvm, float64_t B)
virtual bool	train (CFeatures *data=NULL)
virtual EClassifierType	get_classifier_type ()
virtual CLabels *	classify (CFeatures *data)
virtual CSVM *	get_presvm ()
virtual float64_t	get_B ()
virtual float64_t	get_train_factor ()
virtual void	set_train_factor (float64_t factor)
virtual const char *	get_name () const
保护成员
virtual bool	is_presvm_sane ()
保护属性
CSVM *	presvm
float64_t	B
float64_t	train_factor

---

构造及析构函数文档

CDomainAdaptationSVM

(

)

default constructor

CDomainAdaptationSVM	(	float64_t	C,
		CKernel *	k,
		CLabels *	lab,
		CSVM *	presvm,
		float64_t	B
	)

constructor

参数:

	C	cost constant C
	k	kernel
	lab	labels
	presvm	trained SVM to regularize against
	B	trade-off constant B

virtual ~CDomainAdaptationSVM

(

)

[virtual]

destructor

---

成员函数文档

virtual CLabels* classify

(

CFeatures *

data

)

[virtual]

classify objects

参数:

data

(test)data to be classified

返回:

classified labels

virtual float64_t get_B

(

)

[virtual]

getter for regularization parameter B

返回:

regularization parameter B

virtual EClassifierType get_classifier_type

(

)

[virtual]

get classifier type

返回:

classifier type LIGHT

在文件DomainAdaptationSVM.h第70行定义。

virtual const char* get_name

(

)

const [virtual]

返回:

object name

在文件DomainAdaptationSVM.h第109行定义。

virtual CSVM* get_presvm

(

)

[virtual]

returns SVM that is used as prior information

返回:

presvm

virtual float64_t get_train_factor

(

)

[virtual]

getter for train_factor

返回:

train_factor

void init	(	CSVM *	presvm,
		float64_t	B
	)

init SVM

参数:

	presvm	trained SVM to regularize against
	B	trade-off constant B

virtual bool is_presvm_sane

(

)

[protected, virtual]

check sanity of presvm

返回:

true if sane, throws SG_ERROR otherwise

virtual void set_train_factor

(

float64_t

factor

)

[virtual]

setter for train_factor

virtual bool train

(

CFeatures *

data = NULL

)

[virtual]

train SVM classifier

参数:

data

training data (parameter can be avoided if distance or kernel-based classifiers are used and distance/kernels are initialized with train data)

返回:

whether training was successful

---

成员数据文档

float64_t B [protected]

regularization parameter B

在文件DomainAdaptationSVM.h第154行定义。

CSVM* presvm [protected]

SVM to regularize against

在文件DomainAdaptationSVM.h第150行定义。

float64_t train_factor [protected]

flag to switch off regularization in training

在文件DomainAdaptationSVM.h第158行定义。

---

该类的文档由以下文件生成：

• DomainAdaptationSVM.h