当前位置: 首页>>代码示例 >>用法及示例精选 >>正文


Python PyTorch kron用法及代码示例


本文简要介绍python语言中 torch.kron 的用法。

用法:

torch.kron(input, other, *, out=None) → Tensor

参数

关键字参数

out(Tensor,可选的) -输出张量。如果 None 则忽略。默认值:None

计算 inputother 的克罗内克积,用 表示。

如果 input 张量且 other 张量,则结果将是具有以下条目的 张量:

其中 。如果一个张量的维数比另一个少,则它不会被压缩,直到它具有相同的维数。

支持实值和complex-valued 输入。

注意

如上所述,此函数将两个矩阵的 Kronecker 积的典型定义推广到两个张量。当input 矩阵并且other 矩阵时,结果将是 块矩阵:

其中 input other

例子:

>>> mat1 = torch.eye(2)
>>> mat2 = torch.ones(2, 2)
>>> torch.kron(mat1, mat2)
tensor([[1., 1., 0., 0.],
        [1., 1., 0., 0.],
        [0., 0., 1., 1.],
        [0., 0., 1., 1.]])

>>> mat1 = torch.eye(2)
>>> mat2 = torch.arange(1, 5).reshape(2, 2)
>>> torch.kron(mat1, mat2)
tensor([[1., 2., 0., 0.],
        [3., 4., 0., 0.],
        [0., 0., 1., 2.],
        [0., 0., 3., 4.]])

相关用法


注:本文由纯净天空筛选整理自pytorch.org大神的英文原创作品 torch.kron。非经特殊声明,原始代码版权归原作者所有,本译文未经允许或授权,请勿转载或复制。