rockpool.training.torch_loss

Torch loss functions and regularizers useful for training networks using Torch Modules.

Functions

bounds_cost(params,Β lower_bounds,Β upper_bounds)

Impose a cost on parameters that violate bounds constraints

make_bounds(params)

Convenience function to build a bounds template for a problem

summed_exp_boundary_loss(data[,Β ...])

Compute the summed exponential error of boundary violations of an input.

Classes

ParameterBoundaryRegularizer(*args,Β **kwargs)

Class wrapper for the summed exponential error of boundary violations of an input.