ProbeKind
=========

.. toctree::
   :hidden:

   /autoapi/t3toolbox/backend/fitting/ProbeKind.__post_init__
   /autoapi/t3toolbox/backend/fitting/ProbeKind.precompute
   /autoapi/t3toolbox/backend/fitting/ProbeKind.forward
   /autoapi/t3toolbox/backend/fitting/ProbeKind.transpose
   /autoapi/t3toolbox/backend/fitting/ProbeKind.point_forward
   /autoapi/t3toolbox/backend/fitting/ProbeKind.take

.. py:class:: t3toolbox.backend.fitting.ProbeKind

   Bases: :py:obj:`_WwSample`, :py:obj:`ProbeOutputKind`


   The vector-valued ``probe`` (one free mode per measurement), optionally **per-mode** weighted.

   Mode weighting is the order-0 case of the same residual-weight machinery as the derivative kinds:
   the objective is ``1/2 sum_i |omega_i z_i|^2``, so ``omega`` enters ``sumsq`` (x omega) and
   ``transpose`` (x omega^2) only. Plain probe has no order axis, so the weight is a 1-D ``(d,)``
   per-mode vector.


   .. py:attribute:: name
      :value: 'probe'



   .. py:attribute:: weight
      :type:  t3toolbox.backend.common.typ.Optional[t3toolbox.backend.common.typ.Any]
      :value: None


Methods
-------

.. autoapisummary::

   t3toolbox.backend.fitting.ProbeKind.__post_init__
   t3toolbox.backend.fitting.ProbeKind.precompute
   t3toolbox.backend.fitting.ProbeKind.forward
   t3toolbox.backend.fitting.ProbeKind.transpose
   t3toolbox.backend.fitting.ProbeKind.point_forward
   t3toolbox.backend.fitting.ProbeKind.take