t3toolbox.tucker_tensor_train.t3_corewise_randn#

t3toolbox.tucker_tensor_train.t3_corewise_randn(shape: t3toolbox.backend.common.typ.Tuple[int, Ellipsis], tucker_ranks: t3toolbox.backend.common.typ.Tuple[int, Ellipsis], tt_ranks: t3toolbox.backend.common.typ.Tuple[int, Ellipsis], stack_shape: t3toolbox.backend.common.typ.Tuple[int, Ellipsis] = (), use_jax: bool = False) → TuckerTensorTrain#

Construct a Tucker tensor train with random cores.

Parameters:

  • structure (T3Structure) – Tucker tensor train structure (shape, tucker_ranks, tt_ranks)=((N0,…,N(d-1)), (n0,…,n(d-1)), (1,r1,…,r(d-1),1))).

  • randn (typ.Callable[[..., NDArray]) – Function for creating random arrays. Arguments are a sequence of ints defining the shape of the array. Default: np.random.randn (numpy)

Returns:

Dense tensor represented by x, which has shape (N0, …, N(d-1))

Return type:

NDArray

See also

TuckerTensorTrain, T3Structure

Examples

>>> from t3toolbox import *
>>> import t3toolbox.tucker_tensor_train as t3
>>> shape = (14, 15, 16)
>>> tucker_ranks = (4, 5, 6)
>>> tt_ranks = (1, 3, 2, 1)
>>> stack_shape = (2,3)
>>> x = t3.t3_corewise_randn(shape, tucker_ranks, tt_ranks, stack_shape=stack_shape) # TuckerTensorTrain with random cores
>>> x.uniform_structure == (shape, tucker_ranks, tt_ranks, stack_shape)
True
>>> print(x.tucker_cores[0][0,0,0,0]) # should be random N(0,1)
0.0331003310807162
>>> print(x.tt_cores[0][0,0,0,0,0]) # should be random N(0,1)
-0.10778923886039414