useful_layers
Collection of simple torch.nn.Modules
.
Some of them might be useful when building models using Pytorch Symbolic.
pytorch_symbolic.useful_layers
CallbackLayer
Bases: nn.Module
Layer that returns its inputs, but executes a callable op
on them before returning.
This can be used for debugging or logging purposes. Accepts only one argument.
Example::
x = CallbackLayer(print)(x)
It does not change anything in x, but prints its value.