ISubGVQA.models.mgat
====================

.. py:module:: ISubGVQA.models.mgat


Classes
-------

.. autoapisummary::

   ISubGVQA.models.mgat.MGAT


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

.. py:class:: MGAT(channels, num_ins, dropout=0.0, heads=4, use_instr=False, masking_thresholds=None, use_topk: bool = False, interpretable_mode: bool = True, concat_instr: bool = False, use_all_instrs: bool = False, use_global_mask: bool = False, node_classification: bool = False, sampler_type: str = None, sample_k: int = None, nb_samples: int = 1, alpha=1.0, beta=10.0, tau=1.0)

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


   .. py:attribute:: masking_thresholds
      :value: None



   .. py:attribute:: use_global_mask
      :value: False



   .. py:attribute:: node_classification
      :value: False



   .. py:attribute:: heads
      :value: 4



   .. py:attribute:: use_instr
      :value: False



   .. py:attribute:: use_topk
      :value: False



   .. py:attribute:: interpretable_mode
      :value: True



   .. py:attribute:: use_all_instrs
      :value: False



   .. py:attribute:: convs


   .. py:attribute:: x_proj


   .. py:attribute:: bns


   .. py:attribute:: dropout
      :value: 0.0



   .. py:attribute:: node_logits


   .. py:method:: reset_parameters()


   .. py:method:: forward(x, edge_index, instr_vectors, global_language_feats, edge_attr, batch, return_masks=False, explainer=False, explainer_stage=False, expl_bypass_x=False)