t3_squash_tails
===============

.. py:function:: t3toolbox.backend.t3_operations.t3_squash_tails(data)

   .. code-block:: python

      def t3_squash_tails(
              data: 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+(rLi,ni,rR(i+1))
              ],
      ) -> typ.Tuple[
          typ.Tuple[NDArray, ...],  # tucker_cores, unchanged
          typ.Tuple[NDArray, ...],  # tt_cores with r0=rd=1
      ]:


   Collapse the leading/trailing TT bonds of a T3 to one, without changing the represented tensor.

   The T3-data-level twin of :py:func:`~t3toolbox.backend.ut3_operations.ut3_squash_tails`
   (the Tucker cores are untouched; the TT chain goes through
   :py:func:`~t3toolbox.backend.tt_operations.tt_squash_tails`).