损失为负数

[jerryniu0624]

我改了两个地方：所有的__iter__().next()变成__iter__()._next_data()；[labels[self.get_class(x)] for x in self.train_roots]改成[labels['/'+self.get_class(x)] for x in self.train_roots]，之后损失是负数，这正常吗

[XiHuang1999]

为啥我的改了，next，还是报错？
Traceback (most recent call last):
File J:\代码\PyCharmPythonFiles\FSCF-SSL-2023-main\FSCF-SSL-2023-main\FSCF-SSL.py:882
loss_fsl = crossEntropy(logits, query_labels.cuda())
File E:\Softwares\anacoda\envs\letr\lib\site-packages\torch\nn\modules\module.py:1194 in _call_impl
return forward_call(*input, **kwargs)
File E:\Softwares\anacoda\envs\letr\lib\site-packages\torch\nn\modules\loss.py:1174 in forward
return F.cross_entropy(input, target, weight=self.weight,
File E:\Softwares\anacoda\envs\letr\lib\site-packages\torch\nn\functional.py:3026 in cross_entropy
return torch._C._nn.cross_entropy_loss(input, target, weight, _Reduction.get_enum(reduction), ignore_index, label_smoothing)
RuntimeError: "nll_loss_forward_reduce_cuda_kernel_2d_index" not implemented for 'Int'