SHOGUN  4.0.0
所有成员列表 | Public 成员函数 | Public 属性 | Protected 成员函数 | Protected 属性
CDeepBeliefNetwork类 参考

详细描述

A Deep Belief Network.

A Deep Belief Network [Hinton, 2006] is a multilayer probabilistic generative models. It consists of hidden layers and visible layers. The top hidden layer and the layer below it form a Restricted Boltzmann Machine. The rest of connections in the network are directed connections that go from a hidden layer into a visible layer or another hidden layer.

The network can be pre-trained by treating it as a stack of RBMs. Each hidden layer along with the layer below it form an RBM. Each RBM is then trained using (persistent) contrastive divergence. Pre-training often provides a good initialization for the network's parameters.

After pre-training, the parameters can be fine-tuned using a variant of the wake-sleep algorithm [Hinton, 2006].

The DBN can be used to initialize the parameters of a neural network using convert_to_neural_network().

Samples can be drawn from the model by starting with a random state for the top hidden layer, performing some steps of Gibbs sampling in the top RBM to obtain the states of the top hidden layer and then using those to infer the states of the lower layers using a down-pass.

Steps for using the DBN class:

在文件 DeepBeliefNetwork.h91 行定义.

类 CDeepBeliefNetwork 继承关系图:
Inheritance graph
[图例]

Public 成员函数

 CDeepBeliefNetwork ()
 
 CDeepBeliefNetwork (int32_t num_visible_units, ERBMVisibleUnitType unit_type=RBMVUT_BINARY)
 
virtual ~CDeepBeliefNetwork ()
 
virtual void add_hidden_layer (int32_t num_units)
 
virtual void initialize (float64_t sigma=0.01)
 
virtual void set_batch_size (int32_t batch_size)
 
virtual void pre_train (CDenseFeatures< float64_t > *features)
 
virtual void pre_train (int32_t index, CDenseFeatures< float64_t > *features)
 
virtual void train (CDenseFeatures< float64_t > *features)
 
virtual CDenseFeatures< float64_t > * transform (CDenseFeatures< float64_t > *features, int32_t i=-1)
 
virtual CDenseFeatures< float64_t > * sample (int32_t num_gibbs_steps=1, int32_t batch_size=1)
 
virtual void reset_chain ()
 
virtual CNeuralNetworkconvert_to_neural_network (CNeuralLayer *output_layer=NULL, float64_t sigma=0.01)
 
virtual SGMatrix< float64_tget_weights (int32_t index, SGVector< float64_t > p=SGVector< float64_t >())
 
virtual SGVector< float64_tget_biases (int32_t index, SGVector< float64_t > p=SGVector< float64_t >())
 
virtual const char * get_name () const
 
virtual CSGObjectshallow_copy () const
 
virtual CSGObjectdeep_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="", int32_t param_version=Version::get_version_parameter())
 
virtual bool load_serializable (CSerializableFile *file, const char *prefix="", int32_t param_version=Version::get_version_parameter())
 
DynArray< TParameter * > * load_file_parameters (const SGParamInfo *param_info, int32_t file_version, CSerializableFile *file, const char *prefix="")
 
DynArray< TParameter * > * load_all_file_parameters (int32_t file_version, int32_t current_version, CSerializableFile *file, const char *prefix="")
 
void map_parameters (DynArray< TParameter * > *param_base, int32_t &base_version, DynArray< const SGParamInfo * > *target_param_infos)
 
void set_global_io (SGIO *io)
 
SGIOget_global_io ()
 
void set_global_parallel (Parallel *parallel)
 
Parallelget_global_parallel ()
 
void set_global_version (Version *version)
 
Versionget_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 CSGObjectclone ()
 

Public 属性

SGVector< int32_t > pt_cd_num_steps
 
SGVector< bool > pt_cd_persistent
 
SGVector< bool > pt_cd_sample_visible
 
SGVector< float64_tpt_l2_coefficient
 
SGVector< float64_tpt_l1_coefficient
 
SGVector< int32_t > pt_monitoring_interval
 
SGVector< int32_t > pt_monitoring_method
 
SGVector< int32_t > pt_max_num_epochs
 
SGVector< int32_t > pt_gd_mini_batch_size
 
SGVector< float64_tpt_gd_learning_rate
 
SGVector< float64_tpt_gd_learning_rate_decay
 
SGVector< float64_tpt_gd_momentum
 
int32_t cd_num_steps
 
int32_t monitoring_interval
 
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
 
SGIOio
 
Parallelparallel
 
Versionversion
 
Parameterm_parameters
 
Parameterm_model_selection_parameters
 
Parameterm_gradient_parameters
 
ParameterMapm_parameter_map
 
uint32_t m_hash
 

Protected 成员函数

virtual void down_step (int32_t index, SGVector< float64_t > params, SGMatrix< float64_t > input, SGMatrix< float64_t > result, bool sample_states=true)
 
virtual void up_step (int32_t index, SGVector< float64_t > params, SGMatrix< float64_t > input, SGMatrix< float64_t > result, bool sample_states=true)
 
virtual void wake_sleep (SGMatrix< float64_t > data, CRBM *top_rbm, SGMatrixList< float64_t > sleep_states, SGMatrixList< float64_t > wake_states, SGMatrixList< float64_t > psleep_states, SGMatrixList< float64_t > pwake_states, SGVector< float64_t > gen_params, SGVector< float64_t > rec_params, SGVector< float64_t > gen_gradients, SGVector< float64_t > rec_gradients)
 
virtual TParametermigrate (DynArray< TParameter * > *param_base, const SGParamInfo *target)
 
virtual void one_to_one_migration_prepare (DynArray< TParameter * > *param_base, const SGParamInfo *target, TParameter *&replacement, TParameter *&to_migrate, char *old_name=NULL)
 
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

ERBMVisibleUnitType m_visible_units_type
 
int32_t m_num_layers
 
CDynamicArray< int32_t > * m_layer_sizes
 
SGMatrixList< float64_tm_states
 
int32_t m_batch_size
 
SGVector< float64_tm_params
 
int32_t m_num_params
 
SGVector< int32_t > m_bias_index_offsets
 
SGVector< int32_t > m_weights_index_offsets
 
float64_t m_sigma
 

构造及析构函数说明

default constructor

在文件 DeepBeliefNetwork.cpp51 行定义.

CDeepBeliefNetwork ( int32_t  num_visible_units,
ERBMVisibleUnitType  unit_type = RBMVUT_BINARY 
)

Creates a network with one layer of visible units

参数
num_visible_unitsNumber of visible units
unit_typeType of visible units

在文件 DeepBeliefNetwork.cpp56 行定义.

~CDeepBeliefNetwork ( )
virtual

在文件 DeepBeliefNetwork.cpp65 行定义.

成员函数说明

void add_hidden_layer ( int32_t  num_units)
virtual

Adds a layer of hidden units. The layer is connected to the layer that was added directly before it.

参数
num_unitsNumber of hidden units

在文件 DeepBeliefNetwork.cpp70 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.cpp1243 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.cpp1360 行定义.

CNeuralNetwork * convert_to_neural_network ( CNeuralLayer output_layer = NULL,
float64_t  sigma = 0.01 
)
virtual

Converts the DBN into a neural network with the same structure and parameters. The visible layer in the DBN is converted into a CNeuralInputLayer object, and the hidden layers are converted into CNeuralLogisticLayer objects. An output layer can also be stacked on top of the last hidden layer

参数
output_layerAn output layer
sigmaStandard deviation of the gaussian used to initialize the parameters of the output layer
返回
Neural network inititialized using the DBN

在文件 DeepBeliefNetwork.cpp360 行定义.

CSGObject * deep_copy ( ) const
virtualinherited

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

在文件 SGObject.cpp200 行定义.

void down_step ( int32_t  index,
SGVector< float64_t params,
SGMatrix< float64_t input,
SGMatrix< float64_t result,
bool  sample_states = true 
)
protectedvirtual

Computes the states of some layer using the states of the layer above it

在文件 DeepBeliefNetwork.cpp393 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.cpp1264 行定义.

SGVector< float64_t > get_biases ( int32_t  index,
SGVector< float64_t p = SGVector<float64_t>() 
)
virtual

Returns the bias vector of layer i

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

在文件 DeepBeliefNetwork.cpp558 行定义.

SGIO * get_global_io ( )
inherited

get the io object

返回
io object

在文件 SGObject.cpp237 行定义.

Parallel * get_global_parallel ( )
inherited

get the parallel object

返回
parallel object

在文件 SGObject.cpp278 行定义.

Version * get_global_version ( )
inherited

get the version object

返回
version object

在文件 SGObject.cpp291 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.cpp1135 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.cpp1159 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.cpp1172 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 DeepBeliefNetwork.h214 行定义.

SGMatrix< float64_t > get_weights ( int32_t  index,
SGVector< float64_t p = SGVector<float64_t>() 
)
virtual

Returns the weights matrix between layer i and i+1

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

在文件 DeepBeliefNetwork.cpp547 行定义.

void initialize ( float64_t  sigma = 0.01)
virtual

Initializes the DBN

参数
sigmaStandard deviation of the gaussian used to initialize the weights

在文件 DeepBeliefNetwork.cpp76 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.cpp297 行定义.

DynArray< TParameter * > * load_all_file_parameters ( int32_t  file_version,
int32_t  current_version,
CSerializableFile file,
const char *  prefix = "" 
)
inherited

maps all parameters of this instance to the provided file version and loads all parameter data from the file into an array, which is sorted (basically calls load_file_parameter(...) for all parameters and puts all results into a sorted array)

参数
file_versionparameter version of the file
current_versionversion from which mapping begins (you want to use Version::get_version_parameter() for this in most cases)
filefile to load from
prefixprefix for members
返回
(sorted) array of created TParameter instances with file data

在文件 SGObject.cpp704 行定义.

DynArray< TParameter * > * load_file_parameters ( const SGParamInfo param_info,
int32_t  file_version,
CSerializableFile file,
const char *  prefix = "" 
)
inherited

loads some specified parameters from a file with a specified version The provided parameter info has a version which is recursively mapped until the file parameter version is reached. Note that there may be possibly multiple parameters in the mapping, therefore, a set of TParameter instances is returned

参数
param_infoinformation of parameter
file_versionparameter version of the file, must be <= provided parameter version
filefile to load from
prefixprefix for members
返回
new array with TParameter instances with the attached data

在文件 SGObject.cpp545 行定义.

bool load_serializable ( CSerializableFile file,
const char *  prefix = "",
int32_t  param_version = Version::get_version_parameter() 
)
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
param_version(optional) a parameter version different to (this is mainly for testing, better do not use)
返回
TRUE if done, otherwise FALSE

在文件 SGObject.cpp374 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.

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

在文件 SGObject.cpp1062 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.

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

在文件 SGObject.cpp1057 行定义.

void map_parameters ( DynArray< TParameter * > *  param_base,
int32_t &  base_version,
DynArray< const SGParamInfo * > *  target_param_infos 
)
inherited

Takes a set of TParameter instances (base) with a certain version and a set of target parameter infos and recursively maps the base level wise to the current version using CSGObject::migrate(...). The base is replaced. After this call, the base version containing parameters should be of same version/type as the initial target parameter infos. Note for this to work, the migrate methods and all the internal parameter mappings have to match

参数
param_baseset of TParameter instances that are mapped to the provided target parameter infos
base_versionversion of the parameter base
target_param_infosset of SGParamInfo instances that specify the target parameter base

在文件 SGObject.cpp742 行定义.

TParameter * migrate ( DynArray< TParameter * > *  param_base,
const SGParamInfo target 
)
protectedvirtualinherited

creates a new TParameter instance, which contains migrated data from the version that is provided. The provided parameter data base is used for migration, this base is a collection of all parameter data of the previous version. Migration is done FROM the data in param_base TO the provided param info Migration is always one version step. Method has to be implemented in subclasses, if no match is found, base method has to be called.

If there is an element in the param_base which equals the target, a copy of the element is returned. This represents the case when nothing has changed and therefore, the migrate method is not overloaded in a subclass

参数
param_baseset of TParameter instances to use for migration
targetparameter info for the resulting TParameter
返回
a new TParameter instance with migrated data from the base of the type which is specified by the target parameter

在文件 SGObject.cpp949 行定义.

void one_to_one_migration_prepare ( DynArray< TParameter * > *  param_base,
const SGParamInfo target,
TParameter *&  replacement,
TParameter *&  to_migrate,
char *  old_name = NULL 
)
protectedvirtualinherited

This method prepares everything for a one-to-one parameter migration. One to one here means that only ONE element of the parameter base is needed for the migration (the one with the same name as the target). Data is allocated for the target (in the type as provided in the target SGParamInfo), and a corresponding new TParameter instance is written to replacement. The to_migrate pointer points to the single needed TParameter instance needed for migration. If a name change happened, the old name may be specified by old_name. In addition, the m_delete_data flag of to_migrate is set to true. So if you want to migrate data, the only thing to do after this call is converting the data in the m_parameter fields. If unsure how to use - have a look into an example for this. (base_migration_type_conversion.cpp for example)

参数
param_baseset of TParameter instances to use for migration
targetparameter info for the resulting TParameter
replacement(used as output) here the TParameter instance which is returned by migration is created into
to_migratethe only source that is used for migration
old_namewith this parameter, a name change may be specified

在文件 SGObject.cpp889 行定义.

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

在文件 SGObject.cpp263 行定义.

void pre_train ( CDenseFeatures< float64_t > *  features)
virtual

Pre-trains the DBN as a stack of RBMs

参数
featuresInput features. Should have as many features as the number of visible units in the DBN

在文件 DeepBeliefNetwork.cpp149 行定义.

void pre_train ( int32_t  index,
CDenseFeatures< float64_t > *  features 
)
virtual

Pre-trains a single RBM

参数
indexIndex of the RBM
featuresInput features. Should have as many features as the total number of visible units in the DBN

在文件 DeepBeliefNetwork.cpp159 行定义.

void print_modsel_params ( )
inherited

prints all parameter registered for model selection and their type

在文件 SGObject.cpp1111 行定义.

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

prints registered parameters out

参数
prefixprefix for members

在文件 SGObject.cpp309 行定义.

void reset_chain ( )
virtual

Resets the state of the markov chain used for sampling

在文件 DeepBeliefNetwork.cpp352 行定义.

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

Draws samples from the marginal distribution of the visible units. The sampling starts from the values DBN's internal state and result of the sampling is stored there too.

参数
num_gibbs_stepsNumber of Gibbs sampling steps for the top RBM.
batch_sizeNumber of samples to be drawn. A seperate chain is used for each sample
返回
Sampled states of the visible units

在文件 DeepBeliefNetwork.cpp333 行定义.

bool save_serializable ( CSerializableFile file,
const char *  prefix = "",
int32_t  param_version = Version::get_version_parameter() 
)
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
param_version(optional) a parameter version different to (this is mainly for testing, better do not use)
返回
TRUE if done, otherwise FALSE

在文件 SGObject.cpp315 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.

CKernel 重载.

在文件 SGObject.cpp1072 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.

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.cpp1067 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 DeepBeliefNetwork.cpp135 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp42 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp47 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp52 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp57 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp62 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp67 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp72 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp77 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp82 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp87 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp92 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp97 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp102 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp107 行定义.

void set_generic ( )
inherited

在文件 SGObject.cpp112 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.cpp230 行定义.

void set_global_parallel ( Parallel parallel)
inherited

set the parallel object

参数
parallelparallel object to use

在文件 SGObject.cpp243 行定义.

void set_global_version ( Version version)
inherited

set the version object

参数
versionversion object to use

在文件 SGObject.cpp284 行定义.

CSGObject * shallow_copy ( ) const
virtualinherited

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

CGaussianKernel 重载.

在文件 SGObject.cpp194 行定义.

void train ( CDenseFeatures< float64_t > *  features)
virtual

Trains the DBN using the variant of the wake-sleep algorithm described in [A Fast Learning Algorithm for Deep Belief Nets, Hinton, 2006].

参数
featuresInput features. Should have as many features as the total number of visible units in the DBN

在文件 DeepBeliefNetwork.cpp210 行定义.

CDenseFeatures< float64_t > * transform ( CDenseFeatures< float64_t > *  features,
int32_t  i = -1 
)
virtual

Applies the DBN as a features transformation

Forward-propagates the input features through the DBN and returns the Mean activations of the \( i^{th} \) hidden layer

参数
featuresInput features. Should have as many features as the number of visible units in the DBN
iIndex of the hidden layer. If -1, the activations of the last hidden layer is returned
返回
Mean activations of the \( i^{th} \) hidden layer

在文件 DeepBeliefNetwork.cpp316 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.cpp304 行定义.

void up_step ( int32_t  index,
SGVector< float64_t params,
SGMatrix< float64_t input,
SGMatrix< float64_t result,
bool  sample_states = true 
)
protectedvirtual

Computes the states of some layer using the states of the layer below it

在文件 DeepBeliefNetwork.cpp443 行定义.

void update_parameter_hash ( )
virtualinherited

Updates the hash of current parameter combination

在文件 SGObject.cpp250 行定义.

void wake_sleep ( SGMatrix< float64_t data,
CRBM top_rbm,
SGMatrixList< float64_t sleep_states,
SGMatrixList< float64_t wake_states,
SGMatrixList< float64_t psleep_states,
SGMatrixList< float64_t pwake_states,
SGVector< float64_t gen_params,
SGVector< float64_t rec_params,
SGVector< float64_t gen_gradients,
SGVector< float64_t rec_gradients 
)
protectedvirtual

Computes the gradients using the wake-sleep algorithm

在文件 DeepBeliefNetwork.cpp473 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 wake-sleep training. Default value is 1.

在文件 DeepBeliefNetwork.h306 行定义.

float64_t gd_learning_rate

Gradient descent learning rate for wake-sleep training, defualt value 0.1

在文件 DeepBeliefNetwork.h325 行定义.

float64_t gd_learning_rate_decay

Gradient descent learning rate decay for wake-sleep training. The 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 DeepBeliefNetwork.h332 行定义.

int32_t gd_mini_batch_size

Size of the mini-batch used during gradient descent wake-sleep training, If 0 full-batch training is performed Default value is 0

在文件 DeepBeliefNetwork.h322 行定义.

float64_t gd_momentum

gradient descent momentum multiplier for wake-sleep training

default value is 0.9

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

在文件 DeepBeliefNetwork.h342 行定义.

SGIO* io
inherited

io

在文件 SGObject.h496 行定义.

int32_t m_batch_size
protected

Number of train/test cases the network is currently dealing with

在文件 DeepBeliefNetwork.h358 行定义.

SGVector<int32_t> m_bias_index_offsets
protected

Index at which the bias of each layer is stored in the parameters vector

在文件 DeepBeliefNetwork.h367 行定义.

Parameter* m_gradient_parameters
inherited

parameters wrt which we can compute gradients

在文件 SGObject.h511 行定义.

uint32_t m_hash
inherited

Hash of parameter values

在文件 SGObject.h517 行定义.

CDynamicArray<int32_t>* m_layer_sizes
protected

Size of each layer

在文件 DeepBeliefNetwork.h352 行定义.

Parameter* m_model_selection_parameters
inherited

model selection parameters

在文件 SGObject.h508 行定义.

int32_t m_num_layers
protected

Number of layers

在文件 DeepBeliefNetwork.h349 行定义.

int32_t m_num_params
protected

Number of parameters

在文件 DeepBeliefNetwork.h364 行定义.

ParameterMap* m_parameter_map
inherited

map for different parameter versions

在文件 SGObject.h514 行定义.

Parameter* m_parameters
inherited

parameters

在文件 SGObject.h505 行定义.

SGVector<float64_t> m_params
protected

Parameters of the network

在文件 DeepBeliefNetwork.h361 行定义.

float64_t m_sigma
protected

Standard deviation of the gaussian used to initialize the parameters

在文件 DeepBeliefNetwork.h376 行定义.

SGMatrixList<float64_t> m_states
protected

States of each layer

在文件 DeepBeliefNetwork.h355 行定义.

ERBMVisibleUnitType m_visible_units_type
protected

Type of the visible units

在文件 DeepBeliefNetwork.h346 行定义.

SGVector<int32_t> m_weights_index_offsets
protected

Index at which the weights of each hidden layer is stored in the parameters vector

在文件 DeepBeliefNetwork.h372 行定义.

int32_t max_num_epochs

Maximum number of iterations over the training set during wake-sleep training. Defualt value is 1

在文件 DeepBeliefNetwork.h316 行定义.

int32_t monitoring_interval

Number of weight updates between each evaluation of the reconstruction error during wake-sleep training. Default value is 10.

在文件 DeepBeliefNetwork.h311 行定义.

Parallel* parallel
inherited

parallel

在文件 SGObject.h499 行定义.

SGVector<int32_t> pt_cd_num_steps

CRBM::cd_num_steps for pre-training each RBM. Default value is 1 for all RBMs

在文件 DeepBeliefNetwork.h246 行定义.

SGVector<bool> pt_cd_persistent

CRBM::cd_persistent for pre-training each RBM. Default value is true for all RBMs

在文件 DeepBeliefNetwork.h251 行定义.

SGVector<bool> pt_cd_sample_visible

CRBM::cd_sample_visible for pre-training each RBM. Default value is false for all RBMs

在文件 DeepBeliefNetwork.h256 行定义.

SGVector<float64_t> pt_gd_learning_rate

CRBM::gd_learning_rate for pre-training each RBM. Default value is 0.1 for all RBMs

在文件 DeepBeliefNetwork.h291 行定义.

SGVector<float64_t> pt_gd_learning_rate_decay

CRBM::gd_learning_rate_decay for pre-training each RBM. Default value is 1.0 for all RBMs

在文件 DeepBeliefNetwork.h296 行定义.

SGVector<int32_t> pt_gd_mini_batch_size

CRBM::gd_mini_batch_size for pre-training each RBM. Default value is 0 for all RBMs

在文件 DeepBeliefNetwork.h286 行定义.

SGVector<float64_t> pt_gd_momentum

CRBM::gd_momentum for pre-training each RBM. Default value is 0.9 for all RBMs

在文件 DeepBeliefNetwork.h301 行定义.

SGVector<float64_t> pt_l1_coefficient

CRBM::l1_coefficient for pre-training each RBM. Default value is 0.0 for all RBMs

在文件 DeepBeliefNetwork.h266 行定义.

SGVector<float64_t> pt_l2_coefficient

CRBM::l2_coefficient for pre-training each RBM. Default value is 0.0 for all RBMs

在文件 DeepBeliefNetwork.h261 行定义.

SGVector<int32_t> pt_max_num_epochs

CRBM::max_num_epochs for pre-training each RBM. Default value is 1 for all RBMs

在文件 DeepBeliefNetwork.h281 行定义.

SGVector<int32_t> pt_monitoring_interval

CRBM::monitoring_interval for pre-training each RBM. Default value is 10 for all RBMs

在文件 DeepBeliefNetwork.h271 行定义.

SGVector<int32_t> pt_monitoring_method

CRBM::monitoring_method for pre-training each RBM. Default value is RBMMM_RECONSTRUCTION_ERROR for all RBMs

在文件 DeepBeliefNetwork.h276 行定义.

Version* version
inherited

version

在文件 SGObject.h502 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