quimb.experimental.belief_propagation.mps1bp
============================================

.. py:module:: quimb.experimental.belief_propagation.mps1bp


Classes
-------

.. autoapisummary::

   quimb.experimental.belief_propagation.mps1bp.MPS1BP


Module Contents
---------------

.. py:class:: MPS1BP(tn, max_bond, layer_tags, site_tags=None, *, cutoff=0.0, method='srcmps', fit_messages=True, compress_opts=None, damping=0.0, update='sequential', normalize='L2', distance='L2', local_convergence=True, optimize='auto-hq', **kwargs)

   Bases: :py:obj:`quimb.tensor.belief_propagation.bp_common.BeliefPropagationCommon`


   Matrix product state 1-norm lazy belief propagation.

   :param tn: The tensor network to perform belief propagation on. It should have a
              'graph + 1' structure, i.e. some geometry (given by `site_tags`)
              repeated in layers (given by `layer_tags`).
   :type tn: TensorNetwork
   :param max_bond: The maximum bond dimension to use when compressing the message update
                    tensor network.
   :type max_bond: int
   :param layer_tags: The tags which identify the layers of the tensor network. Each tensor
                      should have exactly one of these tags.
   :type layer_tags: sequence[str, ...]
   :param site_tags: The tags which identify the sites of the tensor network. If None, these
                     are assumed to be default `tn.site_tags`.
   :type site_tags: sequence[str, ...], optional
   :param cutoff: The cutoff to use when compressing the message update tensor network.
                  Default is 0.0.
   :type cutoff: float, optional
   :param method: The method to use when compressing the message update tensor network.
                  See `qtn.tensor_network_1d_compress` for options.
   :type method: str, optional
   :param fit_messages: Whether to use the current message as a fit target when using 'fit'
                        or 'srcmps' compression methods. Default is True.
   :type fit_messages: bool, optional
   :param compress_opts: Additional options to pass to `qtn.tensor_network_1d_compress`.
   :type compress_opts: dict, optional
   :param optimize: The optimization strategy to use when contracting tensor networks.
   :type optimize: str, optional
   :param kwargs: Additional keyword arguments to pass to the base
                  `BeliefPropagationCommon` class.


   .. py:attribute:: local_convergence
      :value: True



   .. py:attribute:: touched


   .. py:attribute:: layer_tags


   .. py:attribute:: max_bond


   .. py:attribute:: compress_opts


   .. py:attribute:: fit_messages
      :value: True



   .. py:attribute:: backend
      :value: 'numpy'



   .. py:attribute:: optimize
      :value: 'auto-hq'



   .. py:attribute:: messages


   .. py:method:: get_message_tn(a, b) -> quimb.tensor.TensorNetwork

      Get the (uncompressed) tensor network describing the current message
      from site `a` to site `b`.



   .. py:method:: compute_message(a, b) -> quimb.tensor.TensorNetwork

      Compute the (compressed) tensor network message from site `a` to
      site `b`.



   .. py:method:: iterate(tol=5e-06)

      Perform one round of message passing.



   .. py:method:: contract(strip_exponent=False, check_zero=True, **kwargs)

      Contract the target tensor network, via the belief propagation
      approximation, using the current messages.

      :param strip_exponent: Whether to return the mantissa and exponent separately.
      :type strip_exponent: bool, optional
      :param check_zero: Whether to check for zero values and return zero early.
      :type check_zero: bool, optional