SHOGUN  4.1.0
 全部 类 命名空间 文件 函数 变量 类型定义 枚举 枚举值 友元 宏定义 组 页
所有成员列表 | Public 成员函数 | Public 属性 | Protected 成员函数 | Protected 属性 | 友元
CRBM类 参考

详细描述

A Restricted Boltzmann Machine.

An RBM is an energy based probabilistic model. It consists of two groups of variables: the visible variables \( v \) and the hidden variables \( h \). The key assumption that RBMs make is that the hidden units are conditionally independent given the visible units, and vice versa.

The energy function for RBMs with binary visible units is defined as:

\[ E(v,h) = - b^T v - c^T h - h^T Wv \]

and for RBMs with gaussian (linear) visible units:

\[ E(v,h) = v^T v - b^T v - c^T h - h^T Wv \]

where \( b \) is the bias vector for the visible units, \( c \) is the bias vector for the hidden units, and \( W \) is the weight matrix.

The probability distribution is defined through the energy fucntion as:

\[ P(v,h) = \frac{exp(-E(v,h))}{\sum_{v,h} exp(-E(v,h))} \]

The above definitions along with the independence assumptions result in the following conditionals:

\[ P(h=1|v) = \frac{1}{1+exp(-Wv-c)} \quad \text{for binary hidden units} \]

\[ P(v=1|h) = \frac{1}{1+exp(-W^T h-b)} \quad \text{for binary visible units} \]

\[ P(v|h) \sim \mathcal{N} (W^T h + b,1) \quad \text{for gaussian visible units} \]

Note that when using gaussian visible units, the inputs should be normalized to have zero mean and unity standard deviation.

This class supports having multiple types of visible units in the same RBM. The visible units are divided into groups where each group can have its own type. The hidden units however are just one group of binary units.

Samples can be drawn from the model using Gibbs sampling.

Training is done using contrastive divergence [Hinton, 2002] or persistent contrastive divergence [Tieleman, 2008] (default).

Training progress can be monitored using the reconstruction error (default), which is the average squared difference between a training batch and the RBM's reconstruction of it. The reconstruction is generated using one step of gibbs sampling. Progress can also be monitored using the pseudo-log-likelihood which is an approximation to the log-likelihood. However, this is currently only supported for binary visible units.

The rows of the visible_state matrix are divided into groups, one for each group of visible units. For example, if we have 3 groups of visible units: group 0 with 10 units, group 1 with 5 units, and group 2 with 6 units, the states of group 0 will be stored in visible_state[0:10,:], the states of group 1 will stored in visible_state[10:15,:], and the states of group 2 will be stored in visible_state[15:21,:]. Note that the groups are numbered by the order in which they where added to the RBM using add_visible_group()

在文件 RBM.h 第 123 行定义.

类 CRBM 继承关系图:
Inheritance graph
[图例]

Public 成员函数

 CRBM ()
 
 CRBM (int32_t num_hidden)
 
 CRBM (int32_t num_hidden, int32_t num_visible, ERBMVisibleUnitType visible_unit_type=RBMVUT_BINARY)
 
virtual ~CRBM ()
 
virtual void add_visible_group (int32_t num_units, ERBMVisibleUnitType unit_type)
 
virtual void initialize_neural_network (float64_t sigma=0.01)
 
virtual void set_batch_size (int32_t batch_size)
 
virtual void train (CDenseFeatures< float64_t > *features)
 
virtual void sample (int32_t num_gibbs_steps=1, int32_t batch_size=1)
 
virtual CDenseFeatures
< float64_t > * 		sample_group (int32_t V, int32_t num_gibbs_steps=1, int32_t batch_size=1)
 
virtual void sample_with_evidence (int32_t E, CDenseFeatures< float64_t > *evidence, int32_t num_gibbs_steps=1)
 
virtual CDenseFeatures
< float64_t > * 		sample_group_with_evidence (int32_t V, int32_t E, CDenseFeatures< float64_t > *evidence, int32_t num_gibbs_steps=1)
 
virtual void reset_chain ()
 
virtual float64_t free_energy (SGMatrix< float64_t > visible, SGMatrix< float64_t > buffer=SGMatrix< float64_t >())
 
virtual void free_energy_gradients (SGMatrix< float64_t > visible, SGVector< float64_t > gradients, bool positive_phase=true, SGMatrix< float64_t > hidden_mean_given_visible=SGMatrix< float64_t >())
 
virtual void contrastive_divergence (SGMatrix< float64_t > visible_batch, SGVector< float64_t > gradients)
 
virtual float64_t reconstruction_error (SGMatrix< float64_t > visible, SGMatrix< float64_t > buffer=SGMatrix< float64_t >())
 
virtual float64_t pseudo_likelihood (SGMatrix< float64_t > visible, SGMatrix< float64_t > buffer=SGMatrix< float64_t >())
 
virtual CDenseFeatures
< float64_t > * 		visible_state_features ()
 
virtual SGVector< float64_t > get_parameters ()
 
virtual SGMatrix< float64_t > get_weights (SGVector< float64_t > p=SGVector< float64_t >())
 
virtual SGVector< float64_t > get_hidden_bias (SGVector< float64_t > p=SGVector< float64_t >())
 
virtual SGVector< float64_t > get_visible_bias (SGVector< float64_t > p=SGVector< float64_t >())
 
virtual int32_t get_num_parameters ()
 
virtual const char * get_name () const
 
virtual CSGObject * shallow_copy () const
 
virtual CSGObject * deep_copy () const
 
virtual bool is_generic (EPrimitiveType *generic) const
 
template<class T >
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
template<>
void set_generic ()
 
void unset_generic ()
 
virtual void print_serializable (const char *prefix="")
 
virtual bool save_serializable (CSerializableFile *file, const char *prefix="")
 
virtual bool load_serializable (CSerializableFile *file, const char *prefix="")
 
void set_global_io (SGIO *io)
 
SGIO * get_global_io ()
 
void set_global_parallel (Parallel *parallel)
 
Parallel * get_global_parallel ()
 
void set_global_version (Version *version)
 
Version * get_global_version ()
 
SGStringList< char > get_modelsel_names ()
 
void print_modsel_params ()
 
char * get_modsel_param_descr (const char *param_name)
 
index_t get_modsel_param_index (const char *param_name)
 
void build_gradient_parameter_dictionary (CMap< TParameter *, CSGObject * > *dict)
 
virtual void update_parameter_hash ()
 
virtual bool parameter_hash_changed ()
 
virtual bool equals (CSGObject *other, float64_t accuracy=0.0, bool tolerant=false)
 
virtual CSGObject * clone ()
 

Public 属性

int32_t cd_num_steps
 
bool cd_persistent
 
bool cd_sample_visible
 
float64_t l2_coefficient
 
float64_t l1_coefficient
 
int32_t monitoring_interval
 
ERBMMonitoringMethod monitoring_method
 
int32_t max_num_epochs
 
int32_t gd_mini_batch_size
 
float64_t gd_learning_rate
 
float64_t gd_learning_rate_decay
 
float64_t gd_momentum
 
SGMatrix< float64_t > hidden_state
 
SGMatrix< float64_t > visible_state
 
SGIO * io
 
Parallel * parallel
 
Version * version
 
Parameter * m_parameters
 
Parameter * m_model_selection_parameters
 
Parameter * m_gradient_parameters
 
uint32_t m_hash
 

Protected 成员函数

virtual void mean_hidden (SGMatrix< float64_t > visible, SGMatrix< float64_t > result)
 
virtual void mean_visible (SGMatrix< float64_t > hidden, SGMatrix< float64_t > result)
 
virtual void sample_hidden (SGMatrix< float64_t > mean, SGMatrix< float64_t > result)
 
virtual void sample_visible (SGMatrix< float64_t > mean, SGMatrix< float64_t > result)
 
virtual void sample_visible (int32_t index, SGMatrix< float64_t > mean, SGMatrix< float64_t > result)
 
virtual void load_serializable_pre () throw (ShogunException)
 
virtual void load_serializable_post () throw (ShogunException)
 
virtual void save_serializable_pre () throw (ShogunException)
 
virtual void save_serializable_post () throw (ShogunException)
 

Protected 属性

int32_t m_num_hidden
 
int32_t m_num_visible
 
int32_t m_batch_size
 
int32_t m_num_visible_groups
 
CDynamicArray< int32_t > * m_visible_group_types
 
CDynamicArray< int32_t > * m_visible_group_sizes
 
CDynamicArray< int32_t > * m_visible_state_offsets
 
int32_t m_num_params
 
SGVector< float64_t > m_params
 

友元

class CDeepBeliefNetwork
 

构造及析构函数说明

CRBM ( )

default constructor

在文件 RBM.cpp 第 44 行定义.

CRBM ( int32_t  num_hidden)

Constructs an RBM with no visible units. The visible units can be added later using add_visible_group()

参数
num_hiddenNumber of hidden units

在文件 RBM.cpp 第 49 行定义.

CRBM ( int32_t  num_hidden,
int32_t  num_visible,
ERBMVisibleUnitType  visible_unit_type = RBMVUT_BINARY 
)

Constructs an RBM with a single group of visible units

参数
num_hiddenNumber of hidden units
num_visibleNumber of visible units
visible_unit_typeType of the visible units

在文件 RBM.cpp 第 55 行定义.

~CRBM ( )
virtual

在文件 RBM.cpp 第 63 行定义.

成员函数说明

void add_visible_group ( int32_t  num_units,
ERBMVisibleUnitType  unit_type 
)
virtual

Adds a group of visible units to the RBM

参数
num_unitsNumber of visible units
visible_unit_typeType of the visible units

在文件 RBM.cpp 第 70 行定义.

void build_gradient_parameter_dictionary ( CMap< TParameter *, CSGObject * > *  dict)
inherited

Builds a dictionary of all parameters in SGObject as well of those of SGObjects that are parameters of this object. Dictionary maps parameters to the objects that own them.

参数
dictdictionary of parameters to be built.

在文件 SGObject.cpp 第 597 行定义.

CSGObject * clone ( )
virtualinherited

Creates a clone of the current object. This is done via recursively traversing all parameters, which corresponds to a deep copy. Calling equals on the cloned object always returns true although none of the memory of both objects overlaps.

返回
an identical copy of the given object, which is disjoint in memory. NULL if the clone fails. Note that the returned object is SG_REF'ed

在文件 SGObject.cpp 第 714 行定义.

void contrastive_divergence ( SGMatrix< float64_t >  visible_batch,
SGVector< float64_t >  gradients 
)
virtual

Computes the gradients using contrastive divergence

参数
visible_batchStates of the visible units
gradientsArray in which the results are stored. Length get_num_parameters()

在文件 RBM.cpp 第 356 行定义.

CSGObject * deep_copy ( ) const
virtualinherited

A deep copy. All the instance variables will also be copied.

在文件 SGObject.cpp 第 198 行定义.

bool equals ( CSGObject *  other,
float64_t  accuracy = 0.0,
bool  tolerant = false 
)
virtualinherited

Recursively compares the current SGObject to another one. Compares all registered numerical parameters, recursion upon complex (SGObject) parameters. Does not compare pointers!

May be overwritten but please do with care! Should not be necessary in most cases.

参数
otherobject to compare with
accuracyaccuracy to use for comparison (optional)
tolerantallows linient check on float equality (within accuracy)
返回
true if all parameters were equal, false if not

在文件 SGObject.cpp 第 618 行定义.

float64_t free_energy ( SGMatrix< float64_t >  visible,
SGMatrix< float64_t >  buffer = SGMatrix<float64_t>() 
)
virtual

Computes the average free energy on a given batch of visible unit states.

The free energy for a vector \( v \) is defined as:

\[ F(v) = - log(\sum_h exp(-E(v,h)) \]

which yields the following (in vectorized form):

\[ F(v) = -b^T v - \sum log(1+exp(Wv+c)) \quad \text{for binary visible units}\]

\[ F(v) = \frac{1}{2} v^T v - b^T v - \sum log(1+exp(Wv+c)) \quad \text{for gaussian visible units}\]

参数
visibleStates of the visible units
bufferA matrix of size num_hidden*batch_size. used as a buffer during computation. If not given, a new matrix is allocated and used as a buffer.
返回
Average free energy over the given batch

在文件 RBM.cpp 第 273 行定义.

void free_energy_gradients ( SGMatrix< float64_t >  visible,
SGVector< float64_t >  gradients,
bool  positive_phase = true,
SGMatrix< float64_t >  hidden_mean_given_visible = SGMatrix<float64_t>() 
)
virtual

Computes the gradients of the free energy function with respect to the RBM's parameters

参数
visibleStates of the visible units
gradientsArray in which the results are stored. Length get_num_parameters()
positive_phaseIf true, the result vector is reset to zero and the gradients are added to it with a positive sign. If false, the result vector is not reset and the gradients are added to it with a negative sign. This is useful during contrastive divergence.
hidden_mean_given_visibleMeans of the hidden states given the visible states. If not given, means will be computed by calling mean_hidden()

在文件 RBM.cpp 第 319 行定义.

SGIO * get_global_io ( )
inherited

get the io object

返回
io object

在文件 SGObject.cpp 第 235 行定义.

Parallel * get_global_parallel ( )
inherited

get the parallel object

返回
parallel object

在文件 SGObject.cpp 第 277 行定义.

Version * get_global_version ( )
inherited

get the version object

返回
version object

在文件 SGObject.cpp 第 290 行定义.

SGVector< float64_t > get_hidden_bias ( SGVector< float64_t >  p = SGVector<float64_t>())
virtual

Returns the bias vector of the hidden units

参数
pIf specified, the bias vector is extracted from it instead of m_params

在文件 RBM.cpp 第 581 行定义.

SGStringList< char > get_modelsel_names ( )
inherited
返回
vector of names of all parameters which are registered for model selection

在文件 SGObject.cpp 第 498 行定义.

char * get_modsel_param_descr ( const char *  param_name)
inherited

Returns description of a given parameter string, if it exists. SG_ERROR otherwise

参数
param_namename of the parameter
返回
description of the parameter

在文件 SGObject.cpp 第 522 行定义.

index_t get_modsel_param_index ( const char *  param_name)
inherited

Returns index of model selection parameter with provided index

参数
param_namename of model selection parameter
返回
index of model selection parameter with provided name, -1 if there is no such

在文件 SGObject.cpp 第 535 行定义.

virtual const char* get_name ( ) const
virtual

Returns the name of the SGSerializable instance. It MUST BE the CLASS NAME without the prefixed `C'.

返回
name of the SGSerializable

实现了 CSGObject.

在文件 RBM.h 第 347 行定义.

virtual int32_t get_num_parameters ( )
virtual

Returns the number of parameters

在文件 RBM.h 第 345 行定义.

virtual SGVector<float64_t> get_parameters ( )
virtual

Returns the parameter vector of the RBM

在文件 RBM.h 第 318 行定义.

SGVector< float64_t > get_visible_bias ( SGVector< float64_t >  p = SGVector<float64_t>())
virtual

Returns the bias vector of the visible units

参数
pIf specified, the bias vector is extracted from it instead of m_params

在文件 RBM.cpp 第 591 行定义.

SGMatrix< float64_t > get_weights ( SGVector< float64_t >  p = SGVector<float64_t>())
virtual

Returns the weights matrix

参数
pIf specified, the weight matrix is extracted from it instead of m_params

在文件 RBM.cpp 第 571 行定义.

void initialize_neural_network ( float64_t  sigma = 0.01)
virtual

Initializes the weights of the RBM. Must be called after all the visible groups have been added, and before the RBM is used.

参数
sigmaStandard deviation of the gaussian used to initialize the weights

在文件 RBM.cpp 第 87 行定义.

bool is_generic ( EPrimitiveType *  generic) const
virtualinherited

If the SGSerializable is a class template then TRUE will be returned and GENERIC is set to the type of the generic.

参数
genericset to the type of the generic if returning TRUE
返回
TRUE if a class template.

在文件 SGObject.cpp 第 296 行定义.

bool load_serializable ( CSerializableFile *  file,
const char *  prefix = "" 
)
virtualinherited

Load this object from file. If it will fail (returning FALSE) then this object will contain inconsistent data and should not be used!

参数
filewhere to load from
prefixprefix for members
返回
TRUE if done, otherwise FALSE

在文件 SGObject.cpp 第 369 行定义.

void load_serializable_post ( )
throw (ShogunException
)
protectedvirtualinherited

Can (optionally) be overridden to post-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::LOAD_SERIALIZABLE_POST is called.

异常
ShogunExceptionwill be thrown if an error occurs.

被 CKernel, CWeightedDegreePositionStringKernel, CList, CAlphabet, CLinearHMM, CGaussianKernel, CInverseMultiQuadricKernel, CCircularKernel , 以及 CExponentialKernel 重载.

在文件 SGObject.cpp 第 426 行定义.

void load_serializable_pre ( )
throw (ShogunException
)
protectedvirtualinherited

Can (optionally) be overridden to pre-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::LOAD_SERIALIZABLE_PRE is called.

异常
ShogunExceptionwill be thrown if an error occurs.

被 CDynamicArray< T >, CDynamicArray< float64_t >, CDynamicArray< float32_t >, CDynamicArray< int32_t >, CDynamicArray< char >, CDynamicArray< bool > , 以及 CDynamicObjectArray 重载.

在文件 SGObject.cpp 第 421 行定义.

void mean_hidden ( SGMatrix< float64_t >  visible,
SGMatrix< float64_t >  result 
)
protectedvirtual

Computes the mean of the hidden states given the visible states

在文件 RBM.cpp 第 451 行定义.

void mean_visible ( SGMatrix< float64_t >  hidden,
SGMatrix< float64_t >  result 
)
protectedvirtual

Computes the mean of the visible states given the hidden states

在文件 RBM.cpp 第 469 行定义.

bool parameter_hash_changed ( )
virtualinherited
返回
whether parameter combination has changed since last update

在文件 SGObject.cpp 第 262 行定义.

void print_modsel_params ( )
inherited

prints all parameter registered for model selection and their type

在文件 SGObject.cpp 第 474 行定义.

void print_serializable ( const char *  prefix = "")
virtualinherited

prints registered parameters out

参数
prefixprefix for members

在文件 SGObject.cpp 第 308 行定义.

float64_t pseudo_likelihood ( SGMatrix< float64_t >  visible,
SGMatrix< float64_t >  buffer = SGMatrix<float64_t>() 
)
virtual

Computes an approximation to the pseudo-likelihood. See this tutorial for more details. Only works with binary visible units

参数
visibleStates of the visible units
bufferA matrix of size num_visible*batch_size. used as a buffer during computation. If not given, a new matrix is allocated and used as a buffer.
returnApproximation to the average pseudo-likelihood over the given batch

在文件 RBM.cpp 第 421 行定义.

float64_t reconstruction_error ( SGMatrix< float64_t >  visible,
SGMatrix< float64_t >  buffer = SGMatrix<float64_t>() 
)
virtual

Computes the average reconstruction error which is defined as:

\[ E = \frac{1}{N} \sum_i (v_i - \widetilde{v})^2 \]

where \( \widetilde{v} \) is computed using one step of gibbs sampling and \( N \) is the batch size

返回
Average reconstruction error over the given batch

在文件 RBM.cpp 第 399 行定义.

void reset_chain ( )
virtual

Resets the state of the markov chain used for sampling, which is stored in the visible_state matrix, to random values

在文件 RBM.cpp 第 266 行定义.

void sample ( int32_t  num_gibbs_steps = 1,
int32_t  batch_size = 1 
)
virtual

Draws samples from the marginal distribution of the visible units using Gibbs sampling. The sampling starts from the values in the RBM's visible_state matrix and result of the sampling is stored there too.

参数
num_gibbs_stepsNumber of Gibbs sampling steps
batch_sizeNumber of samples to be drawn. A seperate chain is used for each sample

在文件 RBM.cpp 第 179 行定义.

CDenseFeatures< float64_t > * sample_group ( int32_t  V,
int32_t  num_gibbs_steps = 1,
int32_t  batch_size = 1 
)
virtual

Draws Samples from \( P(V) \) where \( V \) is one of the visible unit groups. The sampling starts from the values in the RBM's visible_state matrix and result of the sampling is stored there too.

参数
VIndex of the visible unit group to be sampled
num_gibbs_stepsNumber of Gibbs sampling steps
batch_sizeNumber of samples to be drawn. A seperate chain is used for each sample
返回
Sampled states of group V

在文件 RBM.cpp 第 194 行定义.

CDenseFeatures< float64_t > * sample_group_with_evidence ( int32_t  V,
int32_t  E,
CDenseFeatures< float64_t > *  evidence,
int32_t  num_gibbs_steps = 1 
)
virtual

Draws Samples from \( P(V|E=evidence) \) where \( E \) is one of the visible unit groups and \( V \) is another visible unit group. The sampling starts from the values in the RBM's visible_state matrix and result of the sampling is stored there too.

参数
VIndex of the visible unit group to be sampled
EIndex of the evidence visible unit group
evidenceStates of the evidence visible unit group
num_gibbs_stepsNumber of Gibbs sampling steps
返回
Sampled states of group V

在文件 RBM.cpp 第 246 行定义.

void sample_hidden ( SGMatrix< float64_t >  mean,
SGMatrix< float64_t >  result 
)
protectedvirtual

Samples the hidden states according to the provided means

在文件 RBM.cpp 第 520 行定义.

void sample_visible ( SGMatrix< float64_t >  mean,
SGMatrix< float64_t >  result 
)
protectedvirtual

Samples the visible states according to the provided means

在文件 RBM.cpp 第 527 行定义.

void sample_visible ( int32_t  index,
SGMatrix< float64_t >  mean,
SGMatrix< float64_t >  result 
)
protectedvirtual

Samples one group of visible states according to the provided means

在文件 RBM.cpp 第 535 行定义.

void sample_with_evidence ( int32_t  E,
CDenseFeatures< float64_t > *  evidence,
int32_t  num_gibbs_steps = 1 
)
virtual

Draws Samples from \( P(V|E=evidence) \) where \( E \) is one of the visible unit groups and \( V \) is all the visible unit excluding the ones in group \( E \). The sampling starts from the values in the RBM's visible_state matrix and result of the sampling is stored there too.

参数
EIndex of the evidence visible unit group
evidenceStates of the evidence visible unit group
num_gibbs_stepsNumber of Gibbs sampling steps

在文件 RBM.cpp 第 212 行定义.

bool save_serializable ( CSerializableFile *  file,
const char *  prefix = "" 
)
virtualinherited

Save this object to file.

参数
filewhere to save the object; will be closed during returning if PREFIX is an empty string.
prefixprefix for members
返回
TRUE if done, otherwise FALSE

在文件 SGObject.cpp 第 314 行定义.

void save_serializable_post ( )
throw (ShogunException
)
protectedvirtualinherited

Can (optionally) be overridden to post-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::SAVE_SERIALIZABLE_POST is called.

异常
ShogunExceptionwill be thrown if an error occurs.

被 CKernel 重载.

在文件 SGObject.cpp 第 436 行定义.

void save_serializable_pre ( )
throw (ShogunException
)
protectedvirtualinherited

Can (optionally) be overridden to pre-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::SAVE_SERIALIZABLE_PRE is called.

异常
ShogunExceptionwill be thrown if an error occurs.

被 CKernel, CDynamicArray< T >, CDynamicArray< float64_t >, CDynamicArray< float32_t >, CDynamicArray< int32_t >, CDynamicArray< char >, CDynamicArray< bool > , 以及 CDynamicObjectArray 重载.

在文件 SGObject.cpp 第 431 行定义.

void set_batch_size ( int32_t  batch_size)
virtual

Sets the number of train/test cases the RBM will deal with

参数
batch_sizeBatch size

在文件 RBM.cpp 第 96 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 41 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 46 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 51 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 56 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 61 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 66 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 71 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 76 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 81 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 86 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 91 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 96 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 101 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 106 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp 第 111 行定义.

void set_generic ( )
inherited

set generic type to T

void set_global_io ( SGIO *  io)
inherited

set the io object

参数
ioio object to use

在文件 SGObject.cpp 第 228 行定义.

void set_global_parallel ( Parallel *  parallel)
inherited

set the parallel object

参数
parallelparallel object to use

在文件 SGObject.cpp 第 241 行定义.

void set_global_version ( Version *  version)
inherited

set the version object

参数
versionversion object to use

在文件 SGObject.cpp 第 283 行定义.

CSGObject * shallow_copy ( ) const
virtualinherited

A shallow copy. All the SGObject instance variables will be simply assigned and SG_REF-ed.

被 CGaussianKernel 重载.

在文件 SGObject.cpp 第 192 行定义.

void train ( CDenseFeatures< float64_t > *  features)
virtual

Trains the RBM

参数
featuresInput features. Should have as many features as there are visible units in the RBM.

在文件 RBM.cpp 第 108 行定义.

void unset_generic ( )
inherited

unset generic type

this has to be called in classes specializing a template class

在文件 SGObject.cpp 第 303 行定义.

void update_parameter_hash ( )
virtualinherited

Updates the hash of current parameter combination

在文件 SGObject.cpp 第 248 行定义.

virtual CDenseFeatures<float64_t>* visible_state_features ( )
virtual

Returns the states of the visible unit as CDenseFeatures<float64_t>

在文件 RBM.h 第 312 行定义.

友元及相关函数文档

friend class CDeepBeliefNetwork
friend

在文件 RBM.h 第 125 行定义.

类成员变量说明

int32_t cd_num_steps

Number of Gibbs sampling steps performed before each weight update during training. Default value is 1.

在文件 RBM.h 第 373 行定义.

bool cd_persistent

If true, persistent contrastive divergence is used. Default value is true.

在文件 RBM.h 第 377 行定义.

bool cd_sample_visible

If true, the visible units are sampled during contrastive divergence. If false, the visible units are not sampled, and their mean values are used instead. Default value is false

在文件 RBM.h 第 383 行定义.

float64_t gd_learning_rate

gradient descent learning rate, defualt value 0.1

在文件 RBM.h 第 411 行定义.

float64_t gd_learning_rate_decay

gradient descent learning rate decay learning rate is updated at each iteration i according to: alpha(i)=decay*alpha(i-1) default value is 1.0 (no decay)

在文件 RBM.h 第 418 行定义.

int32_t gd_mini_batch_size

size of the mini-batch used during gradient descent training, if 0 full-batch training is performed default value is 0

在文件 RBM.h 第 408 行定义.

float64_t gd_momentum

gradient descent momentum multiplier

default value is 0.9

For more details on momentum, see this paper [Sutskever, 2013]

在文件 RBM.h 第 428 行定义.

SGMatrix<float64_t> hidden_state

States of the hidden units

在文件 RBM.h 第 431 行定义.

SGIO* io
inherited

io

在文件 SGObject.h 第 369 行定义.

float64_t l1_coefficient

L1 Regularization coeff, default value is 0.0

在文件 RBM.h 第 389 行定义.

float64_t l2_coefficient

L2 Regularization coeff, default value is 0.0

在文件 RBM.h 第 386 行定义.

int32_t m_batch_size
protected

Batch size

在文件 RBM.h 第 444 行定义.

Parameter* m_gradient_parameters
inherited

parameters wrt which we can compute gradients

在文件 SGObject.h 第 384 行定义.

uint32_t m_hash
inherited

Hash of parameter values

在文件 SGObject.h 第 387 行定义.

Parameter* m_model_selection_parameters
inherited

model selection parameters

在文件 SGObject.h 第 381 行定义.

int32_t m_num_hidden
protected

Number of hidden units

在文件 RBM.h 第 438 行定义.

int32_t m_num_params
protected

Number of parameters

在文件 RBM.h 第 459 行定义.

int32_t m_num_visible
protected

Number of visible units

在文件 RBM.h 第 441 行定义.

int32_t m_num_visible_groups
protected

Number of visible unit groups

在文件 RBM.h 第 447 行定义.

Parameter* m_parameters
inherited

parameters

在文件 SGObject.h 第 378 行定义.

SGVector<float64_t> m_params
protected

Parameters

在文件 RBM.h 第 462 行定义.

CDynamicArray<int32_t>* m_visible_group_sizes
protected

Size of each visible unit group

在文件 RBM.h 第 453 行定义.

CDynamicArray<int32_t>* m_visible_group_types
protected

Type of each visible unit group

在文件 RBM.h 第 450 行定义.

CDynamicArray<int32_t>* m_visible_state_offsets
protected

Row offsets for accessing the states of each visible unit groups

在文件 RBM.h 第 456 行定义.

int32_t max_num_epochs

maximum number of iterations over the training set. defualt value is 1

在文件 RBM.h 第 402 行定义.

int32_t monitoring_interval

Number of weight updates between each evaluation of the monitoring method. Default value is 10.

在文件 RBM.h 第 394 行定义.

ERBMMonitoringMethod monitoring_method

Monitoring method

在文件 RBM.h 第 397 行定义.

Parallel* parallel
inherited

parallel

在文件 SGObject.h 第 372 行定义.

Version* version
inherited

version

在文件 SGObject.h 第 375 行定义.

SGMatrix<float64_t> visible_state

States of the visible units

在文件 RBM.h 第 434 行定义.

该类的文档由以下文件生成:
SHOGUN 机器学习工具包 - 项目文档