HOOI

# Import statement

from hottbox.algorithms.decomposition import HOOI

class HOOI(init='hosvd', max_iter=50, epsilon=0.01, tol=0.0001, random_state=None, process=(), verbose=False)

Higher Order Orthogonal Iteration Decomposition.

Parameters

initstr

Type of factor matrix initialisation. Available options are hosvd.

processtuple

Specifies the order of modes to be processed. The factor matrices for the missing modes will be set to identity. If empty, then all modes are processed in the consecutive ascending order. Note, initialisation of a factor matrix that corresponds to the mode at the first position is skipped.

max_iterint

Maximum number of iteration

epsilonfloat

Threshold for the relative error of approximation.

tolfloat

Threshold for convergence of factor matrices

random_stateint

verbosebool

If True, enable verbose output

Attributes

cost	(list) A list of relative approximation errors at each iteration of the algorithms.

Methods

copy(self)	Copy of the HOSVD algorithm as a new object
decompose(self, tensor, rank[, keep_meta])	Performs tucker decomposition via Higher Order Orthogonal Iteration (HOOI)

plot

property converged

Checks convergence of the HOOI algorithm.

Returns

bool

copy(self)

Copy of the HOSVD algorithm as a new object

decompose(self, tensor, rank, keep_meta=0)

Performs tucker decomposition via Higher Order Orthogonal Iteration (HOOI)

Parameters

tensorTensor

Multidimensional data to be decomposed

ranktuple

Desired multilinear rank for the given tensor

keep_metaint

Keep meta information about modes of the given tensor. 0 - the output will have default values for the meta data 1 - keep only mode names 2 - keep mode names and indices

Returns

tensor_tkdTensorTKD

Tucker representation of the tensor

property name

Name of the decomposition

Returns

decomposition_namestr