:py:mod:`n2j.models`
====================

.. py:module:: n2j.models


Submodules
----------
.. toctree::
   :titlesonly:
   :maxdepth: 1

   flow/index.rst
   gnn/index.rst
   n2jnet/index.rst


Package Contents
----------------

Classes
~~~~~~~

.. autoapisummary::

   n2j.models.GCNNet
   n2j.models.GATNet
   n2j.models.SageNet
   n2j.models.GravNet
   n2j.models.N2JNet




.. py:class:: GCNNet(in_channels, out_channels, hidden_channels=256, n_layers=3, dropout=0.0, kwargs={})

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

   .. py:method:: forward(data)



.. py:class:: GATNet(in_channels, out_channels, hidden_channels=256, kwargs={}, n_layers=3, dropout=0.0)

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

   .. py:method:: forward(data)



.. py:class:: SageNet(in_channels, out_channels, hidden_channels=256, n_layers=3, dropout=0.0, kwargs={})

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

   .. py:method:: forward(data)



.. py:class:: GravNet(in_channels, out_channels, hidden_channels=256, n_layers=3, dropout=0.0, kwargs={})

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

   .. py:method:: forward(data)



.. py:class:: N2JNet(dim_in, dim_out_local, dim_out_global, dim_local, dim_global, dim_hidden=20, dim_pre_aggr=20, n_iter=20, n_out_layers=5, global_flow=False, dropout=0.0, class_weight=None, use_ss=True, dim_in_meta=2, weight_sum=True)

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

   .. py:method:: forward(data)


   .. py:method:: local_loss(x, data)


   .. py:method:: global_loss(u, data)


   .. py:method:: loss(x, u, data)