tf.invert_permutation

tf.invert_permutation(x, name=None)

tf.math.invert_permutation(x, name=None)

TensorFlow目录总结

入参 / 出参	含义	取值范围
x	A Tensor. Must be one of the following types: int32, int64. 1-D.	其中取值【i】的范围：y[x[i]] = i for i in [0, 1, …, len(x) - 1]； 取值不可重复； 必须从0开始取值
name	A name for the operation (optional).	例如：name = ‘invert_permutation’
Returns	A Tensor. Has the same type as x.

tf.invert_permutation(x, name=None) 置换x数据与索引的关系
tensor x is [3, 4, 0, 2, 1]
invert_permutation(x) ==> [2, 4, 3, 0, 1]

具体使用例如

import tensorflow as tf
# tensorflow 1.14.0
x = [
    1, 2, 4, 5, 3, 0
]
a = tf.math.invert_permutation(x)
with tf.compat.v1.Session() as sess:
    print("x:\n", x)
    print('tf.math.invert_permutation(x):\n', a)
    print('tf.math.invert_permutation(x).run():\n', sess.run(a))

输出结果

x:
 [1, 2, 4, 5, 3, 0]
tf.math.invert_permutation(x):
 Tensor("InvertPermutation:0", shape=(6,), dtype=int32)
tf.math.invert_permutation(x).run():
 [5 0 1 4 2 3]