tensorly.contrib.sparse.decomposition.symmetric_parafac_power_iteration

symmetric_parafac_power_iteration(tensor, rank, n_repeat=10, n_iteration=10, verbose=False)

Symmetric CP Decomposition via Robust Symmetric Tensor Power Iteration

Parameters:
tensortl.tensor

input tensor to decompose, must be symmetric of shape (size, )*order

rankint

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

n_repeatint, default is 10

number of initializations to be tried

n_iterationsint, default is 10

number of power iterations

verbosebool

level of verbosity

Returns:
(weights, factor)
weights1-D tl.tensor of length rank

contains the eigenvalue of each eigenvector

factor2-D tl.tensor of shape (size, rank)

each column corresponds to one eigenvector