:py:mod:`did.dissimilarity`
===========================

.. py:module:: did.dissimilarity

.. autoapi-nested-parse::

   Implements `DIDEstimator` and `NormalizedDIDEstimator` classes to compute the DID dissimilarity
   in an efficient way. Also implements functions for computing `h` and `q` from the precomputed objects.



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

Classes
~~~~~~~

.. autoapisummary::

   did.dissimilarity.DIDEstimator
   did.dissimilarity.NormalizedDIDEstimator



Functions
~~~~~~~~~

.. autoapisummary::

   did.dissimilarity.naive_solve
   did.dissimilarity.compute_h
   did.dissimilarity.compute_q



.. py:class:: DIDEstimator(in_kernel, out_kernel, in_nystrom, out_nystrom)

   Bases: :py:obj:`torch.nn.Module`

   Implementsa class.

   .. py:method:: pre_compute(self, reg_cholesky_in)


   .. py:method:: forward(self, in_f, out_f, mask, in_g, out_g, lambda_)



.. py:class:: NormalizedDIDEstimator(in_kernel, out_kernel, in_nystrom, out_nystrom)

   Bases: :py:obj:`DIDEstimator`

   Implementsa class.

   .. py:method:: forward(self, in_f, out_f, mask, in_g, out_g, lambda_)


   .. py:method:: naive_solve_with_vector(self, lambda_)



.. py:function:: naive_solve(f_hat, g_hat, lambda_)


.. py:function:: compute_h(kernel, coords, x_nystrom_points, r_in_inv, eigenvector)

   Compute h map defined below at image coordinates.


.. py:function:: compute_q(kernel, coords, x_nystrom_points, r_in_inv, eigenvector, f_hat, g_hat, gg_inv)