细数nn.BCELoss与nn.CrossEntropyLoss的区别

Input: (N, *)(N,"htmlcode">

> m = nn.Sigmoid()
> loss = nn.BCELoss()
> input = torch.randn(3, requires_grad=True)
> target = torch.empty(3).random_(2)
> output = loss(m(input), target)
> output.backward()


CrossEntropyLoss:


input(N,C) #n 是batch c是类别
target(N)


输入和target 形状是不同的crossEntropy 是自己会做softmax


> loss = nn.CrossEntropyLoss()
> input = torch.randn(3, 5, requires_grad=True)
> target = torch.empty(3, dtype=torch.long).random_(5)
> output = loss(input, target)
> output.backward()


以上这篇细数nn.BCELoss与nn.CrossEntropyLoss的区别就是小编分享给大家的全部内容了，希望能给大家一个参考，也希望大家多多支持。