如何在 TensorFlow 1.10 中将 (2, 1) 张量重复为 (50, 1) 张量

Question

例如，

# x is a tensor
print(x)
[1, 0] 

# after repeating it
print(x)
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]

TensorFlow 1.10中没有tf.repeat所以有没有最好的可替换实现方式？

Answer 1

如果你真的只能使用 Tensorflow 1.10 然后尝试这样的事情：

import tensorflow as tf

x = tf.constant([1, 0])
x = tf.reshape(tf.tile(tf.expand_dims(x, -1), [1, 25]), (50, 1))
print(x)

'''
tf.Tensor(
[1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0
 0 0 0 0 0 0 0 0 0 0 0 0 0], shape=(50, 1), dtype=int32)
'''

Answer 2

你可以用这个

import tensorflow as tf

x = tf.constant([1, 0])
temp = tf.zeros(shape=(25, 2), dtype=tf.int32)

result = tf.reshape(tf.transpose(temp + x), (-1,))

如何在 TensorFlow 1.10 中将 (2, 1) 张量重复为 (50, 1) 张量

how to repeat (2, 1) tensors to (50, 1) tensors in TensorFlow 1.10

python

tensorflow

tensor