Loss functions
Loss functions can be used both to calculate the loss as the last step in a neural network as well as calculate metrics. In either cases they are provided as an argument to the Workout constructor
# L1Loss as a loss
workout = Workout(model, L1Loss())
# BCELoss as a loss function and FocalLoss as a metric
workout = Workout(model, BCELoss(), floss=FocalLoss())Missing docstring for Loss. Check Documenter's build log for details.
Missing docstring for L1Loss. Check Documenter's build log for details.
Missing docstring for L2Loss. Check Documenter's build log for details.
Missing docstring for LNLoss. Check Documenter's build log for details.
Missing docstring for PseudoHuberLoss. Check Documenter's build log for details.
Missing docstring for BCELoss. Check Documenter's build log for details.
Missing docstring for CrossEntropyLoss. Check Documenter's build log for details.
Missing docstring for HingeLoss. Check Documenter's build log for details.
Missing docstring for SquaredHingeLoss. Check Documenter's build log for details.
Missing docstring for FocalLoss. Check Documenter's build log for details.