公有成员 | 保护属性

CLinearHMM类参考


详细描述

The class LinearHMM is for learning Higher Order Markov chains.

While learning the parameters ${\bf \theta}$ in

\begin{eqnarray*} P({\bf x}|{\bf \theta}^\pm)&=&P(x_1, \ldots, x_N|{\bf \theta}^\pm)\\ &=&P(x_1,\ldots,x_{d}|{\bf \theta}^\pm)\prod_{i=d+1}^N P(x_i|x_{i-1},\ldots,x_{i-d},{\bf \theta}^\pm) \end{eqnarray*}

are determined.

A more detailed description can be found in

Durbin et.al, Biological Sequence Analysis -Probabilistic Models of Proteins and Nucleic Acids, 1998

在文件LinearHMM.h39行定义。

继承图,类CLinearHMM
Inheritance graph
[图例]

所有成员的列表。

公有成员

 CLinearHMM (CStringFeatures< uint16_t > *f)
 CLinearHMM (int32_t p_num_features, int32_t p_num_symbols)
virtual ~CLinearHMM ()
virtual bool train (CFeatures *data=NULL)
bool train (const int32_t *indizes, int32_t num_indizes, float64_t pseudo_count)
float64_t get_log_likelihood_example (uint16_t *vector, int32_t len)
float64_t get_likelihood_example (uint16_t *vector, int32_t len)
virtual float64_t get_log_likelihood_example (int32_t num_example)
virtual float64_t get_log_derivative (int32_t num_param, int32_t num_example)
virtual float64_t get_log_derivative_obsolete (uint16_t obs, int32_t pos)
virtual float64_t get_derivative_obsolete (uint16_t *vector, int32_t len, int32_t pos)
virtual int32_t get_sequence_length ()
virtual int32_t get_num_symbols ()
virtual int32_t get_num_model_parameters ()
virtual float64_t get_positional_log_parameter (uint16_t obs, int32_t position)
virtual float64_t get_log_model_parameter (int32_t num_param)
virtual void get_log_transition_probs (float64_t **dst, int32_t *num)
virtual bool set_log_transition_probs (const float64_t *src, int32_t num)
virtual void get_transition_probs (float64_t **dst, int32_t *num)
virtual bool set_transition_probs (const float64_t *src, int32_t num)
virtual const char * get_name () const

保护属性

int32_t sequence_length
int32_t num_symbols
int32_t num_params
float64_ttransition_probs
float64_tlog_transition_probs

构造及析构函数文档

CLinearHMM ( CStringFeatures< uint16_t > *  f  ) 

constructor

参数:
f features to use

在文件LinearHMM.cpp19行定义。

CLinearHMM ( int32_t  p_num_features,
int32_t  p_num_symbols 
)

constructor

参数:
p_num_features number of features
p_num_symbols number of symbols in features

在文件LinearHMM.cpp28行定义。

~CLinearHMM (  )  [virtual]

在文件LinearHMM.cpp36行定义。


成员函数文档

virtual float64_t get_derivative_obsolete ( uint16_t *  vector,
int32_t  len,
int32_t  pos 
) [virtual]

obsolete get one example's derivative

参数:
vector vector
len length
pos position

在文件LinearHMM.h127行定义。

float64_t get_likelihood_example ( uint16_t *  vector,
int32_t  len 
)

get one example's likelihood

参数:
vector the example
len length of vector
返回:
likelihood

在文件LinearHMM.cpp201行定义。

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

get logarithm of one example's derivative's likelihood

参数:
num_param which example's param
num_example which example
返回:
logarithm of example's derivative

实现了CDistribution

在文件LinearHMM.cpp211行定义。

virtual float64_t get_log_derivative_obsolete ( uint16_t  obs,
int32_t  pos 
) [virtual]

obsolete get logarithm of one example's derivative's likelihood

参数:
obs observation
pos position

在文件LinearHMM.h115行定义。

float64_t get_log_likelihood_example ( uint16_t *  vector,
int32_t  len 
)

get logarithm of one example's likelihood

参数:
vector the example
len length of vector
返回:
logarithm of likelihood

在文件LinearHMM.cpp174行定义。

float64_t get_log_likelihood_example ( int32_t  num_example  )  [virtual]

get logarithm of one example's likelihood

参数:
num_example which example
返回:
logarithm of example's likelihood

实现了CDistribution

在文件LinearHMM.cpp184行定义。

virtual float64_t get_log_model_parameter ( int32_t  num_param  )  [virtual]

get logarithm of given model parameter

参数:
num_param which param
返回:
logarithm of given model parameter

实现了CDistribution

在文件LinearHMM.h169行定义。

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

get logarithm of all transition probs

参数:
dst where logarithm of transition probs will be stored
num where number of logarithm of transition probs will be stored

在文件LinearHMM.cpp260行定义。

virtual const char* get_name (  )  const [virtual]
返回:
object name

实现了CSGObject

在文件LinearHMM.h211行定义。

virtual int32_t get_num_model_parameters (  )  [virtual]

get number of model parameters

返回:
number of model parameters

实现了CDistribution

在文件LinearHMM.h150行定义。

virtual int32_t get_num_symbols (  )  [virtual]

get number of symbols in examples

返回:
number of symbols in examples

在文件LinearHMM.h144行定义。

virtual float64_t get_positional_log_parameter ( uint16_t  obs,
int32_t  position 
) [virtual]

get positional log parameter

参数:
obs observation
position position
返回:
positional log parameter

在文件LinearHMM.h158行定义。

virtual int32_t get_sequence_length (  )  [virtual]

get sequence length of each example

返回:
sequence length of each example

在文件LinearHMM.h138行定义。

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

get all transition probs

参数:
dst where transition probs will be stored
num where number of transition probs will be stored

在文件LinearHMM.cpp230行定义。

bool set_log_transition_probs ( const float64_t src,
int32_t  num 
) [virtual]

set logarithm of all transition probs

参数:
src new logarithms of transition probs
num number of logarithms of transition probs
返回:
if setting was successful

在文件LinearHMM.cpp270行定义。

bool set_transition_probs ( const float64_t src,
int32_t  num 
) [virtual]

set all transition probs

参数:
src new transition probs
num number of transition probs
返回:
if setting was successful

在文件LinearHMM.cpp240行定义。

bool train ( const int32_t *  indizes,
int32_t  num_indizes,
float64_t  pseudo_count 
)

alternative train distribution

参数:
indizes indices
num_indizes number of indices
pseudo_count pseudo count
返回:
if training was successful

在文件LinearHMM.cpp111行定义。

bool train ( CFeatures data = NULL  )  [virtual]

estimate LinearHMM 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

实现了CDistribution

在文件LinearHMM.cpp42行定义。


成员数据文档

logarithm of transition probs

在文件LinearHMM.h223行定义。

int32_t num_params [protected]

number of parameters

在文件LinearHMM.h219行定义。

int32_t num_symbols [protected]

number of symbols in examples

在文件LinearHMM.h217行定义。

int32_t sequence_length [protected]

examples' sequence length

在文件LinearHMM.h215行定义。

transition probs

在文件LinearHMM.h221行定义。


该类的文档由以下文件生成:

SHOGUN Machine Learning Toolbox - Documentation