The implementation of the RBM module. More...

Public Types
typedef DataType::elem_type	ElemType

using	NetworkType = RBM< InitializationRuleType, DataType, PolicyType >

Public Member Functions
	RBM (arma::Mat< ElemType > predictors, InitializationRuleType initializeRule, const size_t visibleSize, const size_t hiddenSize, const size_t batchSize=1, const size_t numSteps=1, const size_t negSteps=1, const size_t poolSize=2, const ElemType slabPenalty=8, const ElemType radius=1, const bool persistence=false)
	Initialize all the parameters of the network using initializeRule. More...

double	Evaluate (const arma::Mat< ElemType > ¶meters, const size_t i, const size_t batchSize)
	Evaluate the RBM network with the given parameters. More...

template
std::enable_if< std::is_same< Policy, BinaryRBM >::value, double >::type	FreeEnergy (const arma::Mat< ElemType > &input)
	This function calculates the free energy of the BinaryRBM. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, double >::type	FreeEnergy (const arma::Mat< ElemType > &input)
	This function calculates the free energy of the SpikeSlabRBM. More...

void	Gibbs (const arma::Mat< ElemType > &input, arma::Mat< ElemType > &output, const size_t steps=SIZE_MAX)
	This function does the k-step Gibbs Sampling. More...

void	Gradient (const arma::Mat< ElemType > ¶meters, const size_t i, arma::Mat< ElemType > &gradient, const size_t batchSize)
	Calculates the gradients for the RBM network. More...

DataType const &	HiddenBias () const
	Return the hidden bias of the network. More...

DataType &	HiddenBias ()
	Modify the hidden bias of the network. More...

template
std::enable_if< std::is_same< Policy, BinaryRBM >::value, void >::type	HiddenMean (const InputType &input, DataType &output)
	The function calculates the mean for the hidden layer. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	HiddenMean (const InputType &input, DataType &output)
	The function calculates the mean of the Normal distribution of P(s\|v, h). More...

size_t const &	HiddenSize () const
	Get the hidden size. More...

size_t	NumFunctions () const
	Return the number of separable functions (the number of predictor points). More...

size_t	NumSteps () const
	Return the number of steps of Gibbs Sampling. More...

const arma::Mat< ElemType > &	Parameters () const
	Return the parameters of the network. More...

arma::Mat< ElemType > &	Parameters ()
	Modify the parameters of the network. More...

template
std::enable_if< std::is_same< Policy, BinaryRBM >::value, void >::type	Phase (const InputType &input, DataType &gradient)
	Calculates the gradient of the RBM network on the provided input. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	Phase (const InputType &input, DataType &gradient)
	Calculates the gradient of the RBM network on the provided input. More...

size_t const &	PoolSize () const
	Get the pool size. More...

template
std::enable_if< std::is_same< Policy, BinaryRBM >::value, void >::type	Reset ()

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	Reset ()

template
std::enable_if< std::is_same< Policy, BinaryRBM >::value, void >::type	SampleHidden (const arma::Mat< ElemType > &input, arma::Mat< ElemType > &output)
	This function samples the hidden layer given the visible layer using Bernoulli function. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	SampleHidden (const arma::Mat< ElemType > &input, arma::Mat< ElemType > &output)
	This function samples the slab outputs from the Normal distribution with mean given by: $h_i\alpha^{-1}W_i^T*v$ and variance: $\alpha^{-1}$ . More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	SampleSlab (InputType &slabMean, DataType &slab)
	The function samples from the Normal distribution of P(s\|v, h), where the mean is given by: $h_i\alpha^{-1}W_i^T*v$ and variance is given by: $\alpha^{-1}$ . More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	SampleSpike (InputType &spikeMean, DataType &spike)
	The function samples the spike function using Bernoulli distribution. More...

template
std::enable_if< std::is_same< Policy, BinaryRBM >::value, void >::type	SampleVisible (arma::Mat< ElemType > &input, arma::Mat< ElemType > &output)
	This function samples the visible layer given the hidden layer using Bernoulli function. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	SampleVisible (arma::Mat< ElemType > &input, arma::Mat< ElemType > &output)
	Sample Hidden function samples the slab outputs from the Normal distribution with mean given by: $h_i\alpha^{-1}W_i^T*v$ and variance: $\alpha^{-1}$ . More...

template
void	serialize (Archive &ar, const unsigned int)
	Serialize the model. More...

void	Shuffle ()
	Shuffle the order of function visitation. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	SlabMean (const DataType &visible, DataType &spike, DataType &slabMean)
	The function calculates the mean of Normal distribution of P(s\|v, h), where the mean is given by: $h_i\alpha^{-1}W_i^T*v$ . More...

ElemType const &	SlabPenalty () const
	Get the regularizer associated with slab variables. More...

DataType const &	SpikeBias () const
	Get the regularizer associated with spike variables. More...

DataType &	SpikeBias ()
	Modify the regularizer associated with spike variables. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	SpikeMean (const InputType &visible, DataType &spikeMean)
	The function calculates the mean of the distribution P(h\|v), where mean is given by: $sigm(v^TW_i\alpha_i^{-1}W_i^Tv + b_i)$ . More...

template
double	Train (OptimizerType &optimizer, CallbackType &&... callbacks)
	Train the RBM on the given input data. More...

DataType const &	VisibleBias () const
	Return the visible bias of the network. More...

DataType &	VisibleBias ()
	Modify the visible bias of the network. More...

template
std::enable_if< std::is_same< Policy, BinaryRBM >::value, void >::type	VisibleMean (InputType &input, DataType &output)
	The function calculates the mean for the visible layer. More...

template
std::enable_if< std::is_same< Policy, SpikeSlabRBM >::value, void >::type	VisibleMean (InputType &input, DataType &output)
	The function calculates the mean of the Normal distribution of P(v\|s, h). More...

DataType const &	VisiblePenalty () const
	Get the regularizer associated with visible variables. More...

DataType &	VisiblePenalty ()
	Modify the regularizer associated with visible variables. More...

size_t const &	VisibleSize () const
	Get the visible size. More...

arma::Cube< ElemType > const &	Weight () const
	Get the weights of the network. More...

arma::Cube< ElemType > &	Weight ()
	Modify the weights of the network. More...

Detailed Description

template
class mlpack::ann::RBM< InitializationRuleType, DataType, PolicyType >

The implementation of the RBM module.

A Restricted Boltzmann Machines (RBM) is a generative stochastic artificial neural network that can learn a probability distribution over its set of inputs. RBMs have found applications in dimensionality reduction, classification, collaborative filtering, feature learning and topic modelling. They can be trained in either supervised or unsupervised ways, depending on the task. They are a variant of Boltzmann machines, with the restriction that the neurons must form a bipartite graph.

Template Parameters

InitializationRuleType	Rule used to initialize the network.
DataType	The type of matrix to be used.
PolicyType	The RBM variant to be used (BinaryRBM or SpikeSlabRBM).

Definition at line 38 of file rbm.hpp.

Member Typedef Documentation

◆ ElemType

typedef DataType::elem_type ElemType

Definition at line 42 of file rbm.hpp.

◆ NetworkType

using NetworkType = RBM

Definition at line 41 of file rbm.hpp.

Constructor & Destructor Documentation

◆ RBM()

RBM	(	arma::Mat< ElemType >	predictors,
		InitializationRuleType	initializeRule,
		const size_t	visibleSize,
		const size_t	hiddenSize,
		const size_t	batchSize = `1`,
		const size_t	numSteps = `1`,
		const size_t	negSteps = `1`,
		const size_t	poolSize = `2`,
		const ElemType	slabPenalty = `8`,
		const ElemType	radius = `1`,
		const bool	persistence = `false`
	)

Initialize all the parameters of the network using initializeRule.

Parameters

predictors	Training data to be used.
initializeRule	InitializationRule object for initializing the network parameter.
visibleSize	Number of visible neurons.
hiddenSize	Number of hidden neurons.
batchSize	Batch size to be used for training.
numSteps	Number of Gibbs Sampling steps.
negSteps	Number of negative samples to average negative gradient.
poolSize	Number of hidden neurons to pool together.
slabPenalty	Regulariser of slab variables.
radius	Feasible regions for visible layer samples.
persistence	Indicates whether to use Persistent CD or not.

Member Function Documentation

◆ Evaluate()

double Evaluate	(	const arma::Mat< ElemType > &	parameters,
		const size_t	i,
		const size_t	batchSize
	)

Evaluate the RBM network with the given parameters.

The function is needed for monitoring the progress of the network.

Parameters

parameters	Matrix model parameters.
i	Index of the data point.
batchSize	Variable to store the present number of inputs.

◆ FreeEnergy() [1/2]

std::enable_ifBinaryRBM>::value, double>::type FreeEnergy ( const arma::Mat< ElemType > & input )

This function calculates the free energy of the BinaryRBM.

The free energy is given by: $-b^Tv - \sum_{i=1}^M log(1 + e^{c_j+v^TW_j})$ .

Parameters

input The visible neurons.

◆ FreeEnergy() [2/2]

std::enable_ifSpikeSlabRBM>::value, double>::type FreeEnergy ( const arma::Mat< ElemType > & input )

This function calculates the free energy of the SpikeSlabRBM.

The free energy is given by: $v^t$$\Delta$v - $\sum_{i=1}^N$ $\log{ \sqrt{\frac{(-2\pi)^K}{\prod_{m=1}^{K}(\alpha_i)_m}}}$ - $\sum_{i=1}^N \log(1+\exp( b_i + \sum_{m=1}^k \frac{(v(w_i)_m^t)^2}{2(\alpha_i)_m})$

Parameters

input The visible layer neurons.

◆ Gibbs()

void Gibbs	(	const arma::Mat< ElemType > &	input,
		arma::Mat< ElemType > &	output,
		const size_t	steps = `SIZE_MAX`
	)

This function does the k-step Gibbs Sampling.

Parameters

input	Input to the Gibbs function.
output	Used for storing the negative sample.
steps	Number of Gibbs Sampling steps taken.

◆ Gradient()

void Gradient	(	const arma::Mat< ElemType > &	parameters,
		const size_t	i,
		arma::Mat< ElemType > &	gradient,
		const size_t	batchSize
	)

Calculates the gradients for the RBM network.

Parameters

parameters	The current parameters of the network.
i	Index of the data point.
gradient	Variable to store the present gradient.
batchSize	Variable to store the present number of inputs.

◆ HiddenBias() [1/2]

DataType const& HiddenBias ( ) const

inline

Return the hidden bias of the network.

Definition at line 354 of file rbm.hpp.

◆ HiddenBias() [2/2]

DataType& HiddenBias ( )

inline

Modify the hidden bias of the network.

Definition at line 356 of file rbm.hpp.

◆ HiddenMean() [1/2]

std::enable_ifBinaryRBM>::value, void>::type HiddenMean	(	const InputType &	input,
		DataType &	output
	)

The function calculates the mean for the hidden layer.

Parameters

input	Visible neurons.
output	Hidden neuron activations.

◆ HiddenMean() [2/2]

std::enable_ifSpikeSlabRBM>::value, void>::type HiddenMean	(	const InputType &	input,
		DataType &	output
	)

The function calculates the mean of the Normal distribution of P(s|v, h).

The mean is given by: $h_i*\alpha^{-1}*W_i^T*v$ The variance is given by: $\alpha^{-1}$

Parameters

input	Visible layer neurons.
output	Consists of both the spike samples and slab samples.

◆ HiddenSize()

size_t const& HiddenSize ( ) const

inline

Get the hidden size.

Definition at line 374 of file rbm.hpp.

◆ NumFunctions()

size_t NumFunctions ( ) const

inline

Return the number of separable functions (the number of predictor points).

Definition at line 333 of file rbm.hpp.

◆ NumSteps()

size_t NumSteps ( ) const

inline

Return the number of steps of Gibbs Sampling.

Definition at line 336 of file rbm.hpp.

◆ Parameters() [1/2]

const arma::Mat<ElemType>& Parameters ( ) const

inline

Return the parameters of the network.

Definition at line 339 of file rbm.hpp.

◆ Parameters() [2/2]

arma::Mat<ElemType>& Parameters ( )

inline

Modify the parameters of the network.

Definition at line 341 of file rbm.hpp.

◆ Phase() [1/2]

std::enable_ifBinaryRBM>::value, void>::type Phase	(	const InputType &	input,
		DataType &	gradient
	)

Calculates the gradient of the RBM network on the provided input.

Parameters

input	The provided input data.
gradient	Stores the gradient of the RBM network.

◆ Phase() [2/2]

std::enable_ifSpikeSlabRBM>::value, void>::type Phase	(	const InputType &	input,
		DataType &	gradient
	)

Calculates the gradient of the RBM network on the provided input.

Parameters

input	The provided input data.
gradient	Stores the gradient of the RBM network.

◆ PoolSize()

size_t const& PoolSize ( ) const

inline

Get the pool size.

Definition at line 376 of file rbm.hpp.

References RBM< InitializationRuleType, DataType, PolicyType >::serialize().

◆ Reset() [1/2]

std::enable_ifBinaryRBM>::value, void>::type Reset ( )

◆ Reset() [2/2]

std::enable_ifSpikeSlabRBM>::value, void>::type Reset ( )

◆ SampleHidden() [1/2]

std::enable_ifBinaryRBM>::value, void>::type SampleHidden	(	const arma::Mat< ElemType > &	input,
		arma::Mat< ElemType > &	output
	)

This function samples the hidden layer given the visible layer using Bernoulli function.

Parameters

input	Visible layer input.
output	The sampled hidden layer.

◆ SampleHidden() [2/2]

std::enable_ifSpikeSlabRBM>::value, void>::type SampleHidden	(	const arma::Mat< ElemType > &	input,
		arma::Mat< ElemType > &	output
	)

This function samples the slab outputs from the Normal distribution with mean given by: $h_i*\alpha^{-1}*W_i^T*v$ and variance: $\alpha^{-1}$ .

Parameters

input	Consists of both visible and spike variables.
output	Sampled slab neurons.

◆ SampleSlab()

std::enable_ifSpikeSlabRBM>::value, void>::type SampleSlab	(	InputType &	slabMean,
		DataType &	slab
	)

The function samples from the Normal distribution of P(s|v, h), where the mean is given by: $h_i*\alpha^{-1}*W_i^T*v$ and variance is given by: $\alpha^{-1}$ .

Parameters

slabMean	Mean of the Normal distribution of the slab neurons.
slab	Sampled slab variable from the Normal distribution.

◆ SampleSpike()

std::enable_ifSpikeSlabRBM>::value, void>::type SampleSpike	(	InputType &	spikeMean,
		DataType &	spike
	)

The function samples the spike function using Bernoulli distribution.

Parameters

spikeMean	Indicates P(h\|v).
spike	Sampled binary spike variables.

◆ SampleVisible() [1/2]

std::enable_ifBinaryRBM>::value, void>::type SampleVisible	(	arma::Mat< ElemType > &	input,
		arma::Mat< ElemType > &	output
	)

This function samples the visible layer given the hidden layer using Bernoulli function.

Parameters

input	Hidden layer of the network.
output	The sampled visible layer.

◆ SampleVisible() [2/2]

std::enable_ifSpikeSlabRBM>::value, void>::type SampleVisible	(	arma::Mat< ElemType > &	input,
		arma::Mat< ElemType > &	output
	)

Sample Hidden function samples the slab outputs from the Normal distribution with mean given by: $h_i*\alpha^{-1}*W_i^T*v$ and variance: $\alpha^{-1}$ .

Parameters

input	Hidden layer of the network.
output	The sampled visible layer.

◆ serialize()

void serialize	(	Archive &	ar,
		const unsigned	int
	)

Serialize the model.

Referenced by RBM< InitializationRuleType, DataType, PolicyType >::PoolSize().

◆ Shuffle()

void Shuffle ( )

Shuffle the order of function visitation.

This may be called by the optimizer.

◆ SlabMean()

std::enable_ifSpikeSlabRBM>::value, void>::type SlabMean	(	const DataType &	visible,
		DataType &	spike,
		DataType &	slabMean
	)

The function calculates the mean of Normal distribution of P(s|v, h), where the mean is given by: $h_i*\alpha^{-1}*W_i^T*v$ .

Parameters

visible	The visible layer neurons.
spike	The spike variables from hidden layer.
slabMean	The mean of the Normal distribution of slab neurons.

◆ SlabPenalty()

ElemType const& SlabPenalty ( ) const

inline

Get the regularizer associated with slab variables.

Definition at line 364 of file rbm.hpp.

◆ SpikeBias() [1/2]

DataType const& SpikeBias ( ) const

inline

Get the regularizer associated with spike variables.

Definition at line 359 of file rbm.hpp.

◆ SpikeBias() [2/2]

DataType& SpikeBias ( )

inline

Modify the regularizer associated with spike variables.

Definition at line 361 of file rbm.hpp.

◆ SpikeMean()

std::enable_ifSpikeSlabRBM>::value, void>::type SpikeMean	(	const InputType &	visible,
		DataType &	spikeMean
	)

The function calculates the mean of the distribution P(h|v), where mean is given by: $sigm(v^T*W_i*\alpha_i^{-1}*W_i^T*v + b_i)$ .

Parameters

visible	The visible layer neurons.
spikeMean	Indicates P(h\|v).

◆ Train()

double Train	(	OptimizerType &	optimizer,
		CallbackType &&...	callbacks
	)

Train the RBM on the given input data.

This will use the existing model parameters as a starting point for the optimization. If this is not what you want, then you should access the parameters vector directly with Parameters() and modify it as desired.

Template Parameters

OptimizerType	Type of optimizer to use to train the model.
CallbackTypes	Types of Callback functions.

Parameters

optimizer	Optimizer type.
callbacks	Callback Functions for ensmallen optimizer `OptimizerType`. See https://www.ensmallen.org/docs.html#callback-documentation.

Returns: The final objective of the trained model (NaN or Inf on error).

◆ VisibleBias() [1/2]

DataType const& VisibleBias ( ) const

inline

Return the visible bias of the network.

Definition at line 349 of file rbm.hpp.

◆ VisibleBias() [2/2]

DataType& VisibleBias ( )

inline

Modify the visible bias of the network.

Definition at line 351 of file rbm.hpp.

◆ VisibleMean() [1/2]

std::enable_ifBinaryRBM>::value, void>::type VisibleMean	(	InputType &	input,
		DataType &	output
	)

The function calculates the mean for the visible layer.

Parameters

input	Hidden neurons from the hidden layer of the network.
output	Visible neuron activations.

◆ VisibleMean() [2/2]

std::enable_ifSpikeSlabRBM>::value, void>::type VisibleMean	(	InputType &	input,
		DataType &	output
	)

The function calculates the mean of the Normal distribution of P(v|s, h).

The mean is given by: $\Lambda^{-1} \sum_{i=1}^N W_i * s_i * h_i$

Parameters

input	Consists of both the spike and slab variables.
output	Mean of the of the Normal distribution.

◆ VisiblePenalty() [1/2]

DataType const& VisiblePenalty ( ) const

inline

Get the regularizer associated with visible variables.

Definition at line 367 of file rbm.hpp.

◆ VisiblePenalty() [2/2]

DataType& VisiblePenalty ( )

inline

Modify the regularizer associated with visible variables.

Definition at line 369 of file rbm.hpp.

◆ VisibleSize()

size_t const& VisibleSize ( ) const

inline

Get the visible size.

Definition at line 372 of file rbm.hpp.

◆ Weight() [1/2]

arma::Cube<ElemType> const& Weight ( ) const

inline

Get the weights of the network.

Definition at line 344 of file rbm.hpp.

◆ Weight() [2/2]

arma::Cube<ElemType>& Weight ( )

inline

Modify the weights of the network.

Definition at line 346 of file rbm.hpp.

The documentation for this class was generated from the following file:

/home/jenkins-mlpack/mlpack.org/_src/mlpack-3.4.2/src/mlpack/methods/ann/rbm/rbm.hpp

Public Types

Public Member Functions

Detailed Description

template class mlpack::ann::RBM< InitializationRuleType, DataType, PolicyType >

Member Typedef Documentation

◆ ElemType

◆ NetworkType

Constructor & Destructor Documentation

◆ RBM()

Member Function Documentation

◆ Evaluate()

◆ FreeEnergy() [1/2]

◆ FreeEnergy() [2/2]

◆ Gibbs()

◆ Gradient()

◆ HiddenBias() [1/2]

◆ HiddenBias() [2/2]

◆ HiddenMean() [1/2]

◆ HiddenMean() [2/2]

◆ HiddenSize()

◆ NumFunctions()

◆ NumSteps()

◆ Parameters() [1/2]

◆ Parameters() [2/2]

◆ Phase() [1/2]

◆ Phase() [2/2]

◆ PoolSize()

◆ Reset() [1/2]

◆ Reset() [2/2]

◆ SampleHidden() [1/2]

◆ SampleHidden() [2/2]

◆ SampleSlab()

◆ SampleSpike()

◆ SampleVisible() [1/2]

◆ SampleVisible() [2/2]

◆ serialize()

◆ Shuffle()

◆ SlabMean()

◆ SlabPenalty()

◆ SpikeBias() [1/2]

◆ SpikeBias() [2/2]

◆ SpikeMean()

◆ Train()

◆ VisibleBias() [1/2]

◆ VisibleBias() [2/2]

◆ VisibleMean() [1/2]

◆ VisibleMean() [2/2]

◆ VisiblePenalty() [1/2]

◆ VisiblePenalty() [2/2]

◆ VisibleSize()

◆ Weight() [1/2]

◆ Weight() [2/2]

template
class mlpack::ann::RBM< InitializationRuleType, DataType, PolicyType >