qq_42750193的博客

DataLoader的使用

import torcha = torch.randn(3,4)print(a)print()i, idx = a.sort(dim=1, descending=True)print(i)print(idx)print()j, rank = idx.sort(dim=1)print(rank)结果：tensor([[ 2.3326, 0.0275, -0.0799, 0.4156], [-2.2066, 1.7997, -2.2767, 0.4704],

用法：torch.max(input) → Tensor返回输入tensor中所有元素的最大值import torcha = torch.randn(1,3)prin(a)print(torch.max(a))######tensor([[-1.2492, -0.1698, 2.3036]])tensor(2.3036)形式： torch.max(input, dim, keepdim=False, out=None) -> (Tensor, LongTensor)torch