BCEWithLogitsLoss

BCEWithLogitsLoss(weight=None, reduction="mean", pos_weight=None)

Binary cross entropy over logits: the sigmoid fused with the loss in one stable pass.

The module form of clika_runtime.nn.functional.binary_cross_entropy_with_logits: construct it once with the settings, then call it like a function.

Args: weight: a per-element rescaling weight, broadcast against input; unset weighs 1; None takes the function's own default. reduction: how the elementwise losses reduce. pos_weight: a weight on the positive class, broadcast against input; unset weighs 1; None takes the function's own default.

__init__

__init__(self, weight: 'Tensor | None' = None, reduction: '_Reduction' = 'mean', pos_weight: 'Tensor | None' = None) -> 'None'

Initialize self. See help(type(self)) for accurate signature.

extra_repr

extra_repr(self) -> 'str'

extra_repr() -> str

One line of per-class detail for :meth:__repr__; a layer prints its geometry here (in_features=64, out_features=256).

forward

forward(self, input: 'Tensor', target: 'Tensor') -> 'Tensor'

forward(input, target) -> Tensor

Runs clika_runtime.nn.functional.binary_cross_entropy_with_logits on input, target with the stored settings; returns a new tensor.