t3_tucker_factors_shared

t3toolbox.backend.sharing.t3_tucker_factors_shared(x, sharing, rtol=1e-09)

def t3_tucker_factors_shared(
        x:          typ.Tuple[
            typ.Sequence[NDArray],  # tucker_cores. len=d, elm_shape=stack_shape+(ni, Ni)
            typ.Sequence[NDArray],  # tt_cores.     len=d, elm_shape=stack_shape+(ri, ni, r(i+1))
        ],
        sharing:    typ.Sequence,   # len=d, static; one hashable group label per mode
        rtol:       float = 1e-9,   # relative tolerance on the factor deviation
) -> NDArray:  # bool array, shape = stack_shape (scalar/0-d when unstacked)

True (per stack element) where the Tucker factors are tied within every sharing group.

The boolean form of t3_sharing_residual() (residual <= rtol) – a non-enforcing checker with an explicit tolerance (the backend is check-free; frontend safe-mode sites pair the residual with safety.effective_rtol instead). Reduce with .all() for a single verdict on a stacked x.

Examples

>>> import numpy as np
>>> import t3toolbox.tucker_tensor_train as t3
>>> import t3toolbox.backend.sharing as sharing
>>> np.random.seed(0)
>>> x = t3.TuckerTensorTrain.randn((6, 6, 5), (3, 3, 2), (1, 2, 2, 1))
>>> tk, tt = x.data
>>> print(bool(sharing.t3_tucker_factors_shared(((tk[0], tk[0], tk[2]), tt), (0, 0, 1))))
True
>>> print(bool(sharing.t3_tucker_factors_shared(x.data, (0, 0, 1))))
False

Parameters:

• x (t3toolbox.backend.common.typ.Tuple[t3toolbox.backend.common.typ.Sequence[NDArray], t3toolbox.backend.common.typ.Sequence[NDArray]])

• sharing (t3toolbox.backend.common.typ.Sequence)

• rtol (float)

Return type:

NDArray