如何在 PyTorch 中找到张量的转置？

在本文中，我们将讨论如何在 PyTorch 中找到张量的转置。通过将行更改为列，将列更改为行来获得转置。我们可以使用 transpose() 方法转置张量。以下语法用于查找张量的转置。

Syntax: torch.transpose(input_tens, dim_0, dim_1)

Parameters:

• input_tens : the input tensor that we want to transpose.
• dim_0 : it will use when we want the first dimension to be transposed..
• dim_1 : it will use when we want the second dimension to be transposed.

Return : this method return transpose of input tensor.

示例 1：

下面的程序是为了了解如何找到 2D 张量的转置。

Python

# import torch module
import torch
 
# Define a 2D tensor
tens = torch.tensor([[10, 20, 30],
                     [40, 50, 60],
                     [70, 80, 90]])
 
# display original tensor
print("\n Original Tensor: \n", tens)
 
# find transpose
tens_transpose = torch.transpose(tens, 0, 1)
print("\n Tensor After Transpose: \n", tens_transpose)

输出：

示例 2：

下面的程序是要知道如何求多维张量的转置。

Python

# import torch module
import torch
 
# Define a 2D tensor
tens = torch.tensor([[[1, 2, 3], [4, 5, 6]],
                     [[7, 8, 9], [10, 11, 12]],
                     [[13, 14, 15], [16, 17, 18]]])
# display original tensor
print("\n Original Tensor: \n", tens)
 
# find transpose of multi-dimension tensor
tens_transpose = torch.transpose(tens, 0, 1)
 
# display final result
print("\n Tensor After Transpose: \n", tens_transpose)

输出：