hxtorch.spiking.functional.unterjubel.Unterjubel
-
class
hxtorch.spiking.functional.unterjubel.Unterjubel(*args, **kwargs) Bases:
torch.autograd.function.FunctionUnterjubel hardware observables to allow correct gradient flow
-
__init__(*args, **kwargs) Initialize self. See help(type(self)) for accurate signature.
Methods
backward(ctx, grad_output)Backward the gradient.
forward(ctx, input, input_prime)Returns input_prime instead of input to inject input_prime but direct the gradient to input.
Attributes
-
static
backward(ctx: torch.Tensor, grad_output: torch.Tensor) → Tuple[Optional[torch.Tensor], …] Backward the gradient.
- Parameters
grad_output – The backwarded gradient.
- Returns
Returns simply the back-propagated gradient at first position.
-
static
forward(ctx, input: torch.Tensor, input_prime: torch.Tensor) → torch.Tensor Returns input_prime instead of input to inject input_prime but direct the gradient to input.
- Parameters
input – Input tensor.
input_prime – The returned tensor.
- Returns
Returns the primed tensor. Thereby, this tensor is forwarded while the gradient is directed to to input.
-