complextorch.nn.masked.linear
=============================

.. py:module:: complextorch.nn.masked.linear

.. autoapi-nested-parse::

   Masked Linear / Bilinear (Complex)
   ==================================

   Layers that apply a fixed binary mask to their complex weight at forward
   time. Used to deploy a learned-sparsity pattern at inference.



Classes
-------

.. autoapisummary::

   complextorch.nn.masked.linear.BilinearMasked
   complextorch.nn.masked.linear.LinearMasked


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

.. py:class:: BilinearMasked(in1_features: int, in2_features: int, out_features: int, bias: bool = True, conjugate: bool = True, device=None, dtype: torch.dtype = torch.cfloat)

   Bases: :py:obj:`complextorch.nn.masked.base.MaskedWeightMixin`, :py:obj:`complextorch.nn.masked.base.BaseMasked`


   Complex bilinear with a fixed binary weight mask.

   Initialize internal Module state, shared by both nn.Module and ScriptModule.


   .. py:method:: forward(input1: torch.Tensor, input2: torch.Tensor) -> torch.Tensor


   .. py:attribute:: conjugate
      :value: True



   .. py:attribute:: in1_features


   .. py:attribute:: in2_features


   .. py:attribute:: out_features


   .. py:attribute:: weight


.. py:class:: LinearMasked(in_features: int, out_features: int, bias: bool = True, device=None, dtype: torch.dtype = torch.cfloat)

   Bases: :py:obj:`complextorch.nn.masked.base.MaskedWeightMixin`, :py:obj:`complextorch.nn.masked.base.BaseMasked`


   Complex linear with a fixed binary weight mask.

   Initialize internal Module state, shared by both nn.Module and ScriptModule.


   .. py:method:: forward(input: torch.Tensor) -> torch.Tensor


   .. py:attribute:: in_features


   .. py:attribute:: out_features


   .. py:attribute:: weight