mlpack
Public Member Functions | List of all members
mlpack::kernel::NystroemMethod< KernelType, PointSelectionPolicy > Class Template Reference

Public Member Functions

 NystroemMethod (const arma::mat &data, KernelType &kernel, const size_t rank)
 Create the NystroemMethod object. More...
 
void Apply (arma::mat &output)
 Apply the low-rank factorization to obtain an output matrix G such that K' = G * G^T. More...
 
void GetKernelMatrix (const arma::mat *data, arma::mat &miniKernel, arma::mat &semiKernel)
 Construct the kernel matrix with matrix that contains the selected points. More...
 
void GetKernelMatrix (const arma::Col< size_t > &selectedPoints, arma::mat &miniKernel, arma::mat &semiKernel)
 Construct the kernel matrix with the selected points. More...
 

Constructor & Destructor Documentation

◆ NystroemMethod()

template<typename KernelType , typename PointSelectionPolicy >
mlpack::kernel::NystroemMethod< KernelType, PointSelectionPolicy >::NystroemMethod ( const arma::mat &  data,
KernelType &  kernel,
const size_t  rank 
)

Create the NystroemMethod object.

The constructor here does not really do anything.

Parameters
dataData matrix.
kernelKernel to be used for computation.
rankRank to be used for matrix approximation.

Member Function Documentation

◆ Apply()

template<typename KernelType , typename PointSelectionPolicy >
void mlpack::kernel::NystroemMethod< KernelType, PointSelectionPolicy >::Apply ( arma::mat &  output)

Apply the low-rank factorization to obtain an output matrix G such that K' = G * G^T.

Parameters
outputMatrix to store kernel approximation into.

◆ GetKernelMatrix() [1/2]

template<typename KernelType , typename PointSelectionPolicy >
void mlpack::kernel::NystroemMethod< KernelType, PointSelectionPolicy >::GetKernelMatrix ( const arma::mat *  data,
arma::mat &  miniKernel,
arma::mat &  semiKernel 
)

Construct the kernel matrix with matrix that contains the selected points.

Parameters
dataData matrix pointer.
miniKernelto store the constructed mini-kernel matrix in.
semiKernelto store the constructed semi-kernel matrix in.

◆ GetKernelMatrix() [2/2]

template<typename KernelType , typename PointSelectionPolicy >
void mlpack::kernel::NystroemMethod< KernelType, PointSelectionPolicy >::GetKernelMatrix ( const arma::Col< size_t > &  selectedPoints,
arma::mat &  miniKernel,
arma::mat &  semiKernel 
)

Construct the kernel matrix with the selected points.

Parameters
selectedPointsIndices of selected points.
miniKernelto store the constructed mini-kernel matrix in.
semiKernelto store the constructed semi-kernel matrix in.

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