Hello, I’m having trouble creating a Label Smoothing function for CrossEntropy.
input : (1, 3, 512, 512)
output : (512, 512) # n_class = 12
class LabelSmoothingLoss(nn.Module):
def __init__(self, classes, smoothing=0.0, dim=-1):
super(LabelSmoothingLoss, self).__init__()
self.confidence = 1.0 - smoothing
self.smoothing = smoothing
self.cls = classes
self.dim = dim
def forward(self, pred, target):
pred = pred.log_softmax(dim=self.dim)
with torch.no_grad():
true_dist = torch.zeros_like(pred)
true_dist.fill_(self.smoothing / (self.cls - 1))
true_dist.scatter_(1, target.data.unsqueeze(1), self.confidence)
return torch.mean(torch.sum(-true_dist * pred, dim=self.dim))
In this situation, which part of the code above should I modify?