This class computes SVD using complete incremental batch learning, as described in the following paper: More...

Public Member Functions
	SVDCompleteIncrementalLearning (double u=0.0001, double kw=0, double kh=0)
	Initialize the SVDCompleteIncrementalLearning class with the given parameters. More...

void	HUpdate (const MatType &V, const arma::mat &W, arma::mat &H)
	The update rule for the encoding matrix H. More...

void	Initialize (const MatType &, const size_t)
	Initialize parameters before factorization. More...

void	WUpdate (const MatType &V, arma::mat &W, const arma::mat &H)
	The update rule for the basis matrix W. More...

Private Attributes
size_t	currentItemIndex
	Item index of current entry. More...

size_t	currentUserIndex
	User of index of current entry. More...

double	kh
	Regularization parameter for matrix H. More...

double	kw
	Regularization parameter for matrix W. More...

double	u
	Step count of batch learning. More...

Detailed Description

template<class MatType>
class mlpack::amf::SVDCompleteIncrementalLearning< MatType >

This class computes SVD using complete incremental batch learning, as described in the following paper:

@techreport{ma2008guide,
  title={A Guide to Singular Value Decomposition for Collaborative
      Filtering},
  author={Ma, Chih-Chao},
  year={2008},
  institution={Department of Computer Science, National Taiwan University}
}

This class implements 'Algorithm 3' given in the paper. Complete incremental learning is an extreme case of incremental learning, where feature vectors are updated after looking at each single element in the input matrix (V). This approach differs from incomplete incremental learning where feature vectors are updated after seeing columns of elements in the input matrix.

See also: SVDIncompleteIncrementalLearning

Definition at line 45 of file svd_complete_incremental_learning.hpp.

Constructor & Destructor Documentation

template<class MatType >

mlpack::amf::SVDCompleteIncrementalLearning< MatType >::SVDCompleteIncrementalLearning	(	double	u = `0.0001`,
		double	kw = `0`,
		double	kh = `0`
	)

inline

Initialize the SVDCompleteIncrementalLearning class with the given parameters.

Parameters

u	Step value used in batch learning.
kw	Regularization constant for W matrix.
kh	Regularization constant for H matrix.

Definition at line 56 of file svd_complete_incremental_learning.hpp.

Member Function Documentation

template<class MatType >

void mlpack::amf::SVDCompleteIncrementalLearning< MatType >::HUpdate	(	const MatType &	V,
		const arma::mat &	W,
		arma::mat &	H
	)

inline

The update rule for the encoding matrix H.

The function takes in all the matrices and only changes the value of the H matrix.

Parameters

V	Input matrix to be factorized.
W	Basis matrix.
H	Encoding matrix to be updated.

Definition at line 123 of file svd_complete_incremental_learning.hpp.

References mlpack::amf::SVDCompleteIncrementalLearning< MatType >::currentItemIndex, mlpack::amf::SVDCompleteIncrementalLearning< MatType >::currentUserIndex, mlpack::amf::SVDCompleteIncrementalLearning< MatType >::kh, and mlpack::amf::SVDCompleteIncrementalLearning< MatType >::u.

template<class MatType >

void mlpack::amf::SVDCompleteIncrementalLearning< MatType >::Initialize	(	const MatType &	,
		const size_t
	)

inline

Initialize parameters before factorization.

This function must be called before a new factorization. For this initialization, the input parameters are unnecessary; we are only setting the current element index to 0.

Parameters