公有成员 | 保护属性

CDistribution类参考


详细描述

Base class Distribution from which all methods implementing a distribution are derived.

Distributions are based on some general feature object and have to implement interfaces to

train() - for learning a distribution get_num_model_parameters() - for the total number of model parameters get_log_model_parameter() - for the n-th model parameter (logarithmic) get_log_derivative() - for the partial derivative wrt. to the n-th model parameter get_log_likelihood_example() - for the likelihood for the n-th example

This way methods building on CDistribution, might enumerate over all possible model parameters and obtain the parameter vector and the gradient. This is used to compute e.g. the TOP and Fisher Kernel (cf. CPluginEstimate, CHistogramKernel, CTOPFeatures and CFKFeatures ).

在文件Distribution.h41行定义。

继承图,类CDistribution
Inheritance graph
[图例]

所有成员的列表。

公有成员

 CDistribution ()
virtual ~CDistribution ()
virtual bool train (CFeatures *data=NULL)=0
virtual int32_t get_num_model_parameters ()=0
virtual int32_t get_num_relevant_model_parameters ()
virtual float64_t get_log_model_parameter (int32_t num_param)=0
virtual float64_t get_log_derivative (int32_t num_param, int32_t num_example)=0
virtual float64_t get_log_likelihood_example (int32_t num_example)=0
virtual float64_t get_log_likelihood_sample ()
virtual void get_log_likelihood (float64_t **dst, int32_t *num)
virtual float64_t get_model_parameter (int32_t num_param)
virtual float64_t get_derivative (int32_t num_param, int32_t num_example)
virtual float64_t get_likelihood_example (int32_t num_example)
virtual void set_features (CFeatures *f)
virtual CFeaturesget_features ()
virtual void set_pseudo_count (float64_t pseudo)
virtual float64_t get_pseudo_count ()

保护属性

CFeaturesfeatures
float64_t pseudo_count

构造及析构函数文档

CDistribution (  ) 

default constructor

在文件Distribution.cpp16行定义。

~CDistribution (  )  [virtual]

在文件Distribution.cpp22行定义。


成员函数文档

virtual float64_t get_derivative ( int32_t  num_param,
int32_t  num_example 
) [virtual]

get partial derivative of likelihood function

参数:
num_param partial derivative against which param
num_example which example
返回:
derivative of likelihood function

在文件Distribution.h130行定义。

virtual CFeatures* get_features (  )  [virtual]

get feature vectors

返回:
feature vectors

在文件Distribution.h161行定义。

virtual float64_t get_likelihood_example ( int32_t  num_example  )  [virtual]

compute likelihood for example

参数:
num_example which example
返回:
likelihood for example

在文件Distribution.h141行定义。

virtual float64_t get_log_derivative ( int32_t  num_param,
int32_t  num_example 
) [pure virtual]

get partial derivative of likelihood function (logarithmic)

abstract base method

参数:
num_param derivative against which param
num_example which example
返回:
derivative of likelihood (logarithmic)

CGHMMCHistogramCHMMCLinearHMM内被实现。

void get_log_likelihood ( float64_t **  dst,
int32_t *  num 
) [virtual]

compute log likelihood for each example

参数:
dst where likelihood will be stored
num where number of likelihoods will be stored

在文件Distribution.cpp37行定义。

virtual float64_t get_log_likelihood_example ( int32_t  num_example  )  [pure virtual]

compute log likelihood for example

abstract base method

参数:
num_example which example
返回:
log likelihood for example

CGHMMCHistogramCHMMCLinearHMM内被实现。

float64_t get_log_likelihood_sample (  )  [virtual]

compute log likelihood for whole sample

返回:
log likelihood for whole sample

在文件Distribution.cpp26行定义。

virtual float64_t get_log_model_parameter ( int32_t  num_param  )  [pure virtual]

get model parameter (logarithmic)

abstrac base method

返回:
model parameter (logarithmic)

CGHMMCHistogramCHMMCLinearHMM内被实现。

virtual float64_t get_model_parameter ( int32_t  num_param  )  [virtual]

get model parameter

参数:
num_param which param
返回:
model parameter

在文件Distribution.h119行定义。

virtual int32_t get_num_model_parameters (  )  [pure virtual]

get number of parameters in model

abstract base method

返回:
number of parameters in model

CGHMMCHistogramCHMMCLinearHMM内被实现。

int32_t get_num_relevant_model_parameters (  )  [virtual]

get number of parameters in model that are relevant, i.e. > ALMOST_NEG_INFTY

返回:
number of relevant model parameters

在文件Distribution.cpp50行定义。

virtual float64_t get_pseudo_count (  )  [virtual]

get pseudo count

返回:
pseudo count

在文件Distribution.h177行定义。

virtual void set_features ( CFeatures f  )  [virtual]

set feature vectors

参数:
f new feature vectors

在文件Distribution.h150行定义。

virtual void set_pseudo_count ( float64_t  pseudo  )  [virtual]

set pseudo count

参数:
pseudo new pseudo count

在文件Distribution.h171行定义。

virtual bool train ( CFeatures data = NULL  )  [pure virtual]

learn distribution

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

CGHMMCHistogramCHMMCLinearHMM内被实现。


成员数据文档

CFeatures* features [protected]

feature vectors

在文件Distribution.h181行定义。

float64_t pseudo_count [protected]

pseudo count

在文件Distribution.h183行定义。


该类的文档由以下文件生成:

SHOGUN Machine Learning Toolbox - Documentation