linmult.core.pe
===============

.. py:module:: linmult.core.pe

.. autoapi-nested-parse::

   Sinusoidal positional encoding with optional dropout.



Classes
-------

.. autoapisummary::

   linmult.core.pe.PositionalEncoding


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

.. py:class:: PositionalEncoding(dropout: float = 0.1)

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


   Sinusoidal positional encoding for sequence inputs.

   Adds fixed sinusoidal position encodings to the input tensor, following
   Vaswani et al. (2017). The encoding matrix is computed lazily and cached;
   it is only recomputed when the sequence is longer or the feature dimension
   changes.

   :param dropout: Dropout probability applied after adding the encoding.
                   Defaults to ``0.1``.
   :type dropout: float

   Initialize PositionalEncoding.


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

      Add sinusoidal positional encoding to the input.

      The encoding matrix is rebuilt only when the cached tensor is shorter
      than the current sequence or the feature dimensionality has changed.
      For odd feature dimensions, the cosine slot count is ``floor(F/2)``
      while the sine slot count is ``ceil(F/2)``; the division term is
      sliced accordingly so no index is out of range.

      :param x: Input tensor of shape ``(B, T, F)``.
      :type x: torch.Tensor

      :returns: Encoded tensor of shape ``(B, T, F)`` with dropout applied.
      :rtype: torch.Tensor