fastdev.nn.pointnet2
====================

.. py:module:: fastdev.nn.pointnet2

.. autoapi-nested-parse::

   PointNet++ implementation in PyTorch.

   Adapted from: https://github.com/yanx27/Pointnet_Pointnet2_pytorch





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

.. py:class:: PointNet2Encoder(normal_channel: bool = False, feature_dim: int = 1024, n_points: Tuple[int, int] = (512, 128), n_samples: Tuple[int, int] = (32, 64), radius: Tuple[float, float] = (0.2, 0.4))

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


   .. py:attribute:: normal_channel
      :value: False



   .. py:attribute:: sa1


   .. py:attribute:: sa2


   .. py:attribute:: sa3


   .. py:method:: forward(xyz)


.. py:class:: PointNet2Cls(num_class: int, normal_channel=True)

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


   .. py:attribute:: normal_channel
      :value: True



   .. py:attribute:: sa1


   .. py:attribute:: sa2


   .. py:attribute:: sa3


   .. py:attribute:: fc1


   .. py:attribute:: bn1


   .. py:attribute:: drop1


   .. py:attribute:: fc2


   .. py:attribute:: bn2


   .. py:attribute:: drop2


   .. py:attribute:: fc3


   .. py:attribute:: relu


   .. py:method:: forward(xyz)