澄南澄北的博客

官网是这么介绍CrossEntropyLoss 的: This criterion combines LogSoftmax and NLLLoss in one single class. 它将 LogSoftmax 与 NLLLoss 结合在单个类中. 用相应的函数进行验证: import torch import torch.nn.functional as F N = 32 num_class = 1000 input = torch.randn(N, num_class) targe.