topobench.nn.wrappers.base module

Abstract class that provides an interface to handle the network output.

class ABC

Bases: object

Helper class that provides a standard way to create an ABC using inheritance.

class AbstractWrapper(backbone, **kwargs)

Bases: ABC, Module

Abstract class that provides an interface to handle the network output.

Parameters:

backbonetorch.nn.Module

Backbone model.

**kwargsdict

Additional arguments for the class. It should contain the following keys: - out_channels (int): Number of output channels. - num_cell_dimensions (int): Number of cell dimensions.

__init__(backbone, **kwargs)

abstract forward(batch)

Forward pass for the model.

Parameters:

batchtorch_geometric.data.Data

Batch object containing the batched data.

residual_connection(model_out, batch)

Residual connection for the model.

This method sums, for the embeddings of the cells of any rank, the output of the model with the input embeddings and applies layer normalization.

Parameters:

model_outdict

Dictionary containing the model output.

batchtorch_geometric.data.Data

Batch object containing the batched data.

Returns:

dict

Dictionary containing the updated model output.

abstractmethod(funcobj)

A decorator indicating abstract methods.

Requires that the metaclass is ABCMeta or derived from it. A class that has a metaclass derived from ABCMeta cannot be instantiated unless all of its abstract methods are overridden. The abstract methods can be called using any of the normal ‘super’ call mechanisms. abstractmethod() may be used to declare abstract methods for properties and descriptors.

Usage:

class C(metaclass=ABCMeta):

@abstractmethod def my_abstract_method(self, …):

…