张量扩展tf.tile()

tf.tile(input, multiples,name=None)

输出的值是input重复multiples次

例1：

import tensorflow as tf

inputs = tf.constant([[1, 2, 3], [4, 5, 6]], dtype=tf.float32)
result = tf.tile(inputs, multiples=[2, 2])

with tf.Session() as sess:
    print(inputs.eval())
    print('-------inputs在axes=0,1上复制2次---------')
    print(sess.run(result))

结果1：

[[1. 2. 3.]
 [4. 5. 6.]]
-------inputs在axes=0,1上复制2次---------
[[1. 2. 3. 1. 2. 3.]
 [4. 5. 6. 4. 5. 6.]
 [1. 2. 3. 1. 2. 3.]
 [4. 5. 6. 4. 5. 6.]]

例2：

import tensorflow as tf

inputs = tf.Variable(tf.random_normal(shape=(1, 2, 2)))
result = tf.tile(inputs, multiples=[2, 1, 1])

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    print(inputs.eval())
    print('-------inputs在axes=0上复制2次,1,2维复制1次（也就是原矩阵）---------')
    print(sess.run(result))

结果2：

[[[0.38975003 0.13419515]
  [0.46780697 1.4279771 ]]]
-------inputs在axes=0上复制2次,1,2维复制1次（也就是原矩阵）---------
[[[0.38975003 0.13419515]
  [0.46780697 1.4279771 ]]

 [[0.38975003 0.13419515]
  [0.46780697 1.4279771 ]]]