Tensorflow 初始化给出了所有的 [英] Tensorflow initialization gives all ones

问题描述

张量流 1.12.0

在下面截取的代码中，似乎wrapped_rv_val 和seq_rv_val 应该是等价的，但实际上不是.相反，seq_rv_val 被正确地初始化为随机生成的 init_val 数组，但wrapped_rv_val 被设置为全1.这是怎么回事?

In the code snipped below, it seems that wrapped_rv_val and seq_rv_val should be equivalent, but they are not. Instead, seq_rv_val is correctly initialized to the randomly generated init_val array, but wrapped_rv_val is set to all ones. What's going on here?

import numpy as np
import tensorflow as tf

init_val = np.random.rand(1, 1, 16, 1).astype(np.dtype('float32'))

wrapped_rv = tf.nn.softmax(tf.get_variable('wrapped_rv', initializer=init_val))

var = tf.get_variable('seq_rv', initializer=init_val)
seq_rv = tf.nn.softmax(var, axis=2)

init_op = tf.global_variables_initializer()

with tf.Session() as sess:
    sess.run(init_op)
    wrapped_rv_val = sess.run(wrapped_rv)
    seq_rv_val = sess.run(seq_rv)

    print("seq_rv_val: {0}".format(seq_rv_val.flatten()))
    print("wrapped_rv_val: {0}".format(wrapped_rv_val.flatten())) 

输出:

seq_rv_val: [0.28422353 0.12556878 0.18170598 0.19684952 0.21165217]

seq_rv_val: [0.28422353 0.12556878 0.18170598 0.19684952 0.21165217]

wrapped_rv_val: [1.1. 1. 1. 1.]

wrapped_rv_val: [1. 1. 1. 1. 1.]

推荐答案

事实上，seq_rv_val 和 wrapped_rv_val 都会正确初始化为随机生成的 init_val数组，当您执行以下操作时.

In fact, seq_rv_val and wrapped_rv_val both will be correctly initialized to the randomly generated init_val array when you do the following.

# change
wrapped_rv = tf.nn.softmax(tf.get_variable('wrapped_rv', initializer=init_val))
# to
wrapped_rv = tf.nn.softmax(tf.get_variable('wrapped_rv', initializer=init_val), axis=2)

接下来解释一下为什么wrapped_rv被初始化为1，我们来看一下softmax的公式.

Next I'll explain why wrapped_rv is initialized to 1. Let's look at the formula of softmax.

当您设置axis=2时，分母求和项的数量将为16.但是当您设置axis=-1(默认)时，分母求和项的数量将为1.所以分子和分母是一样的，当你把它设置为axis=-1时，结果是1.你可以运行下面的例子来理解问题.

The number of denominator summation items will be 16 when you set axis=2. But the number of denominator summation items will be 1 when you set axis=-1(default). So the molecule is the same as the denominator and the result is 1 when you set it to axis=-1. You can run the following example to understand the problem.

import tensorflow as tf

y = tf.constant([[1],[0],[1]],dtype=tf.float32)
y1 = tf.constant([[1],[2],[3]],dtype=tf.float32)
y2 = tf.constant([[1],[3],[7]],dtype=tf.float32)

softmax_var1 = tf.nn.softmax(logits=y1)
softmax_var2 = tf.nn.softmax(logits=y2)

with tf.Session() as sess:
    print(sess.run(softmax_var1))
    print(sess.run(softmax_var2))

[[1.]
 [1.]
 [1.]]
[[1.]
 [1.]
 [1.]]

这篇关于Tensorflow 初始化给出了所有的的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！