如何在 PyTorch 中的特定新维度中重复张量 [英] How to repeat tensor in a specific new dimension in PyTorch

问题描述

如果我有一个形状为 [M, N] 的张量 A，我想重复张量 K 次，以便结果 B 具有形状 [M, K, N]并且每个切片 B[:, k, :] 应该与 A 具有相同的数据.这是没有 for 循环的最佳实践.K 可能在另一个维度.

If I have a tensor A which has shape [M, N], I want to repeat the tensor K times so that the result B has shape [M, K, N] and each slice B[:, k, :] should has the same data as A. Which is the best practice without a for loop. K might be in other dimension.

torch.repeat_interleave() 和 tensor.repeat() 似乎不起作用.或者我用错了方法.

torch.repeat_interleave() and tensor.repeat() does not seem to work. Or I am using it in a wrong way.

推荐答案

tensor.repeat 应该适合您的需求，但您需要先插入一个单一维度.为此，我们可以使用 tensor.reshape 或 tensor.unsqueeze.由于 unsqueeze 专门定义为插入单一维度，我们将使用它.

tensor.repeat should suit your needs but you need to insert a unitary dimension first. For this we could use either tensor.reshape or tensor.unsqueeze. Since unsqueeze is specifically defined to insert a unitary dimension we will use that.

B = A.unsqueeze(1).repeat(1, K, 1)

代码说明 A.unsqueeze(1) 将 A 从 [M, N] 变成 [M, 1, N] 和 .repeat(1, K, 1) 沿第二维重复张量 K 次.

Code Description A.unsqueeze(1) turns A from an [M, N] to [M, 1, N] and .repeat(1, K, 1) repeats the tensor K times along the second dimension.

这篇关于如何在 PyTorch 中的特定新维度中重复张量的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！