BCEAfterSigmoidLoss

class BCEAfterSigmoidLoss(reduction: Literal['mean', 'sum'] = 'mean')[source]

Bases: PointwiseLoss

The numerically unstable version of explicit Sigmoid + BCE loss.

Note

The related torch module is torch.nn.BCELoss, but it can not be used interchangeably in PyKEEN because of the extended functionality implemented in PyKEEN’s loss functions.

Initialize the loss.

Parameters:

reduction (Literal['mean', 'sum']) – the reduction, cf. pykeen.nn.modules._Loss

Methods Summary

forward(x, target[, weight])

Calculate the point-wise loss.

Methods Documentation

forward(x: Tensor, target: Tensor, weight: Tensor | None = None) → Tensor[source]

Calculate the point-wise loss.

Parameters:

  • x (Tensor) – The predictions.

  • target (Tensor) – The target values (between 0 and 1).

  • weight (Tensor | None) – The sample weights.

Returns:

The scalar loss value.

Return type:

Tensor