tv_probe_transpose_derivatives_from_sweep
=========================================

.. py:function:: t3toolbox.backend.sampling_derivatives.tv_probe_transpose_derivatives_from_sweep(ztildes, ww, pp, frame, sweep, order, sum_over_probes = False, chunk_size = 100)

   .. code-block:: python

      def tv_probe_transpose_derivatives_from_sweep(
              ztildes:    typ.Sequence[NDArray],  # residual jets, len=d, elm_shape=(order+1,)+W+K+C+(Ni,)
              ww:         typ.Sequence[NDArray],  # probe vectors X,        len=d, elm_shape=W+(Ni,)
              pp:         typ.Sequence[NDArray],  # perturbation vectors P, len=d, elm_shape=W+(Ni,)
              frame:       typ.Tuple[
                  typ.Sequence[NDArray], typ.Sequence[NDArray],
                  typ.Sequence[NDArray], typ.Sequence[NDArray],
              ],                                  # = T3Frame.data = (U, O, P, Q)
              sweep:      typ.Tuple[
                  typ.Sequence[NDArray], typ.Sequence[NDArray],
                  typ.Sequence[NDArray], typ.Sequence[NDArray],
              ],                                  # = tv_precompute_probe_frame_sweep_jets(frame, ww, pp, order)
              order:      int,                    # highest derivative order K
              sum_over_probes: bool = False,      # True: sum the sample stack W (the J^T r back-projection)
              chunk_size: typ.Optional[int] = 100,   # W-chunk size for the gradient assembly; None -> dense. docs/chunking.md
      ) -> typ.Tuple[
          typ.Tuple[NDArray, ...],  # dU_tildes (Tucker variation gradient)
          typ.Tuple[NDArray, ...],  # dG_tildes (TT variation gradient)
      ]:                                          # = T3Variations.data


   Variation gradient of :py:func:`tv_probe_derivatives_transpose` from a precomputed frame
   ``sweep``: the adjoint sweeps (``sigma/tau/dxi_tilde`` jets) + the order-less gradient assembly,
   reusing the frame ``(xi, mu, nu, eta)_jets``. The reuse hook for a fitting inner solve.