Core

class kagu.core.loss.Loss

MSE Loss function is applied to minimize the distance between the prediction and actual input. This MSE loss is applied on the whole snippet.

forward(pred, label)

The loss calculation in the forward function

Parameters:

• pred (torch.Tensor) – The predicted features of the model

• label (torch.Tensor) – The actual inputs features

Returns:

• (torch.Tensor): The prediction loss

• (torch.Tensor): The motion loss.

• (torch.Tensor): The motion weighted loss