UniformTuckerTensorTrain.ones#

static t3toolbox.uniform_tucker_tensor_train.UniformTuckerTensorTrain.ones(shape, stack_shape=(), use_jax=False)#

def ones(
        shape:       Sequence[int],          # (N0,...,N(d-1))
        stack_shape: Sequence[int] = (),
        use_jax:     bool = False,
) -> 'UniformTuckerTensorTrain':

Rank-1 uniform Tucker tensor train representing a tensor full of ones (every real entry == 1).

Examples

>>> import numpy as np
>>> import t3toolbox.uniform_tucker_tensor_train as ut3
>>> x = ut3.UniformTuckerTensorTrain.ones((5, 6, 7), stack_shape=(2,))
>>> print(float(np.linalg.norm(x.to_dense() - np.ones((2, 5, 6, 7)))))
0.0
>>> print(np.asarray(x.tucker_ranks).max(), np.asarray(x.tt_ranks).max())
1 1

Parameters:

shape (Sequence[int])
stack_shape (Sequence[int])
use_jax (bool)

Return type:

UniformTuckerTensorTrain