tensorly.decomposition.sample_khatri_rao

sample_khatri_rao(matrices, n_samples, skip_matrix=None, indices_list=None, return_sampled_rows=False, random_state=None)[source]

Random subsample of the Khatri-Rao product of the given list of matrices

If one matrix only is given, that matrix is directly returned.

Parameters:

matricesndarray list

list of matrices with the same number of columns, i.e.:

for i in len(matrices):
    matrices[i].shape = (n_i, m)

n_samplesint

number of samples to be taken from the Khatri-Rao product

skip_matrixNone or int, optional, default is None

if not None, index of a matrix to skip

indices_listlist, default is None

Contains, for each matrix in matrices, a list of indices of rows to sample. if None, random indices will be created

random_stateNone, int or numpy.random.RandomState

if int, used to set the seed of the random number generator if numpy.random.RandomState, used to generate random_samples

returned_sampled_rowsbool, default is False

if True, also returns a list of the rows sampled from the full khatri-rao product

Returns:

sampled_Khatri_Raondarray

The sampled matricised tensor Khatri-Rao with n_samples rows

indicestuple list

a list of indices sampled for each mode

indices_krint list

list of length n_samples containing the sampled row indices