torch.Tensor.repeat

torch.Tensor.repeat(*sizes) → Tensor

Repeats this tensor along the specified dimensions. 

• Unlike expand(), this function copies the tensor’s data.
• repeat() behaves differently from numpy.repeat, but is more similar to numpy.tile.
• For the operator similar to numpy.repeat, see torch.repeat_interleave().

Parameters

• sizes (torch.Size or int...) – The number of times to repeat this tensor along each dimension

>>> x = torch.tensor([1, 2, 3])
>>> x.repeat(4, 2)
tensor([[ 1,  2,  3,  1,  2,  3],
        [ 1,  2,  3,  1,  2,  3],
        [ 1,  2,  3,  1,  2,  3],
        [ 1,  2,  3,  1,  2,  3]])
>>> x.repeat(4, 2, 1).size()
torch.Size([4, 2, 3])

个人(仅仅是猜测，还没有完全验证)：具体如何复制，可以根据参数sizes，从后面的维度向前面的维度，将x进行复制，也就是首先从后往前，从参数sizes得到x的维度个维度，然后根据得到的这个元组的各个分量复制x,整个过程完成后，再每次从后向前参数sizes，每次增加一个维度，复制x,

x = torch.tensor([[1,2,3,4],[99,88,77,66]])

y = x.repeat((5,1,1))