ISubGVQA.sampling.methods.tensor_utils
======================================

.. py:module:: ISubGVQA.sampling.methods.tensor_utils


Functions
---------

.. autoapisummary::

   ISubGVQA.sampling.methods.tensor_utils.batched_edge_index_to_batched_adj
   ISubGVQA.sampling.methods.tensor_utils.self_defined_softmax
   ISubGVQA.sampling.methods.tensor_utils.weighted_cross_entropy
   ISubGVQA.sampling.methods.tensor_utils.non_merge_coalesce
   ISubGVQA.sampling.methods.tensor_utils.batch_repeat_edge_index


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

.. py:function:: batched_edge_index_to_batched_adj(data: torch_geometric.data.Data)

   Args:
       data: should be the original batch, i.e. without ensembles

   Returns:
       batched edge_index ([0, 0, 0, 1, 1, 1, ...],
                           [0, 0, 1, 0, 1, 1, ...],
                           [1, 2, 1, 0, 0, 2, ...]) with self loops



.. py:function:: self_defined_softmax(scores, mask)

   A specific function

   Args:
       scores: B, N, N, E
       mask: same shape as scores

   Returns:



.. py:function:: weighted_cross_entropy(pred, true)

   Weighted cross-entropy for unbalanced classes.
   https://github.com/rampasek/GraphGPS/blob/main/graphgps/loss/weighted_cross_entropy.py


.. py:function:: non_merge_coalesce(edge_index, edge_attr, edge_weight, num_nodes, is_sorted: bool = False, sort_by_row: bool = True)

.. py:function:: batch_repeat_edge_index(edge_index: torch.Tensor, num_nodes: int, repeats: int)