t3toolbox.backend.tucker_tensor_train.ragged_t3_operations

Functions

to_dense(→ t3toolbox.backend.common.NDArray)	Fully contract a Tucker tensor train to create a dense tensor.
squash_tt_tails(...)	Make leading and trailing TT ranks equal to 1 (r0=rd=1), without changing tensor being represented.
reverse_tt(...)	Reverse a tensor train (no Tucker).
change_tucker_core_shapes(...)	Increase Tucker and/or TT ranks for TT cores using zero padding.
change_tt_core_shapes(...)	Increase Tucker and/or TT ranks for TT cores using zero padding.
t3_unstack(x)	Given multiple stacked T3s, this unstacks them
t3_sum_stack(...)	If this object contains multiple stacked T3s, this sums them.
t3_core_shapes(...)	Determines the shapes of the T3 cores based on the ranks.
t3_to_vector(→ t3toolbox.backend.common.NDArray)	Converts T3 to a 1D vector containing all of the backend entries.
t3_from_vector(...)	Constructs a T3 from a 1D vector containing the backend entries
t3_corewise_randn(, use_jax, ...)	Construct a Tucker tensor train with random cores.
t3_zeros(, use_jax, ...)	Construct a Tucker tensor train of zeros.