tensorly.decomposition.CPPower

class CPPower(rank, n_repeat=10, n_iteration=10, verbose=0)[source]

CP Decomposition via Robust Tensor Power Iteration

Parameters:

tensortl.tensor

input tensor to decompose

rankint

rank of the decomposition (number of rank-1 components)

n_repeatint, default is 10

number of initializations to be tried

n_iterationint, default is 10

number of power iterations

verbosebool

level of verbosity

Returns:

(weights, factors)

weights1-D tl.tensor of length rank

contains the eigenvalue of each eigenvector

factorslist of 2-D tl.tensor of shape (size, rank)

Each column of each factor corresponds to one eigenvector

fit_transform(tensor)[source]

Decompose an input tensor

Parameters:

tensortensorly tensor

input tensor to decompose

Returns:

CPTensor

decomposed tensor