Duplicate log_softmax

[mathmanu]

Looking throughh the source code for torch.nn.CrossEntropyLoss(), it seems to be doing log_softmax inside it. The model is also doing a log_softmax at the end. Isn't this a duplicate? Is it needed?

From model.py:

    def forward(self, x):
        ...

        # flatten for input to fully-connected layer
        x = x.view(x.size(0), -1)
        x = self.fc(x)
        return F.log_softmax(x, dim=1)

From Pytorch source code:

class CrossEntropyLoss(_WeightedLoss):
...
    def __init__(self, weight=None, size_average=None, ignore_index=-100,
                 reduce=None, reduction='elementwise_mean'):
        super(CrossEntropyLoss, self).__init__(weight, size_average, reduce, reduction)
        self.ignore_index = ignore_index

    def forward(self, input, target):
        return F.cross_entropy(input, target, weight=self.weight,
                               ignore_index=self.ignore_index, reduction=self.reduction)

def cross_entropy(input, target, weight=None, size_average=None, ignore_index=-100,
                  reduce=None, reduction='elementwise_mean'):
...
    if size_average is not None or reduce is not None:
        reduction = _Reduction.legacy_get_string(size_average, reduce)
    return nll_loss(log_softmax(input, 1), target, weight, None, ignore_index, None, reduction)

[ChrisHRZ]

same qusetion