tf.transpose()的用法

一、tensorflow官方文档内容

transpose(
    a,
    perm=None,
    name='transpose'
)

Defined in tensorflow/python/ops/array_ops.py.

See the guides: Math > Matrix Math FunctionsTensor Transformations > Slicing and Joining

Transposes a. Permutes the dimensions according to perm.

The returned tensor's dimension i will correspond to the input dimension perm[i]. If perm is not given, it is set to (n-1...0), where n is the rank of the input tensor. Hence by default, this operation performs a regular matrix transpose on 2-D input Tensors.

 

For example:  

# 'x' is [[1 2 3]
#         [4 5 6]]
tf.transpose(x) ==> [[1 4]
                     [2 5]
                     [3 6]]

# Equivalently
tf.transpose(x, perm=[1, 0]) ==> [[1 4]
                                  [2 5]
                                  [3 6]]

# 'perm' is more useful for n-dimensional tensors, for n > 2
# 'x' is   [[[1  2  3]
#            [4  5  6]]
#           [[7  8  9]
#            [10 11 12]]]
# Take the transpose of the matrices in dimension-0
tf.transpose(x, perm=[0, 2, 1]) ==> [[[1  4]
                                      [2  5]
                                      [3  6]]

                                     [[7 10]
                                      [8 11]
                                      [9 12]]]

Args:

  • a: A Tensor.
  • perm: A permutation of the dimensions of a.
  • name: A name for the operation (optional).

Returns:

A transposed Tensor.

 

二、中文翻译

transpose(
    a,
    perm=None,
    name='transpose'
)

Defined in tensorflow/python/ops/array_ops.py.

See the guides: Math > Matrix Math FunctionsTensor Transformations > Slicing and Joining

a的转置是根据 perm 的设定值来进行的。 

返回数组的 dimension(尺寸、维度) i与输入的 perm[i]的维度相一致。如果未给定perm,默认设置为 (n-1...0),这里的 n 值是输入变量的 rank 。因此默认情况下,这个操作执行了一个正规(regular)的2维矩形的转置。

 

例子:

# 'x' is [[1 2 3]
#         [4 5 6]]
tf.transpose(x) ==> [[1 4]
                     [2 5]
                     [3 6]]

# Equivalently(等价于)
tf.transpose(x, perm=[1, 0]) ==> [[1 4]
                                  [2 5]
                                  [3 6]]

# 'perm' is more useful for n-dimensional tensors, for n > 2
# 'x' is   [[[1  2  3]
#            [4  5  6]]
#           [[7  8  9]
#            [10 11 12]]]
# Take the transpose of the matrices in dimension-0
tf.transpose(x, perm=[0, 2, 1]) ==> [[[1  4]
                                      [2  5]
                                      [3  6]]

                                     [[7 10]
                                      [8 11]
                                      [9 12]]]

 

参数:  

  • a: a 是一个张量(Tensor)
  • perm: perm 是 a 维度的置换
  • name:操作的名称(可选).

 

返回值:

   返回的是一个转置的张量。

 

三、解释

tf.transpose(input, [dimension_1, dimenaion_2,..,dimension_n]):这个函数主要适用于交换输入张量的不同维度用的,如果输入张量是二维,就相当是转置。dimension_n是整数,如果张量是三维,就是用0,1,2来表示。这个列表里的每个数对应相应的维度。如果是[2,1,0],就把输入张量的第三维度和第一维度交换。  

 
----------------------------------
参考链接:
  1、 tf.transpose函数的用法: https://i.cnblogs.com/EditPosts.aspx?opt=1
  2、 tensorflow中的不懂得知识点——转置函数 transpose : http://blog.csdn.net/u010417185/article/details/51900441
  3、tensorflow官方文档 https://www.tensorflow.org/versions/r1.3/api_docs/python/tf/transpose
posted @ 2018-02-26 21:37  AI菌  阅读(5792)  评论(0编辑  收藏  举报