ut3_inner_product
=================

.. py:function:: t3toolbox.backend.ut3_linalg.ut3_inner_product(x, y)

   .. code-block:: python

      def ut3_inner_product(x: UT3Data, y: UT3Data) -> NDArray:  # HS inner product, shape=stack_shape


   Hilbert-Schmidt inner product of two uniform Tucker tensor trains. Masks (zero padding) and
   squashes both, absorbs the Tucker cores into the TT cores, then zippers the two trains to a scalar
   via a scan over the modes. Orthogonalization (for stability) is the frontend's job, applied first.