tf.random_normal

Tensorflow系列：tf.random_normal

原创 2017年01月03日 13:40:09

• 标签：
• tensorflow

• 7915

在CNN代码里，可能有这样一句话： W1 = tf.Variable(tf.random_normal([3, 3, 1, 32], stddev=0.01))，对于函数tf.random_normal解释如下：

tf.random_normal

tf.random_normal(shape, mean=0.0, stddev=1.0, dtype=tf.float32, seed=None, name=None)

Outputs random values from a normal distribution.

Args:

shape: A 1-D integer Tensor or Python array. The shape of the output tensor.用一个list表示产出的Tensor的形状

mean: A 0-D Tensor or Python value of type dtype. The mean of the normal distribution.均值

stddev: A 0-D Tensor or Python value of type dtype. The standard deviation of the normal distribution.标准差

dtype: The type of the output.数据类型

seed: A Python integer. Used to create a random seed for the distribution. See set_random_seed for behavior.

name: A name for the operation (optional).

Returns:

A tensor of the specified shape filled with random normal values.

例子：

<pre name="code" class="python">norm = tf.random_normal([2, 3], seed=1234)
sess = tf.Session()
print(sess.run(norm))
print(sess.run(norm))</pre>
<pre></pre>

输出：

[[ 0.51340485 -0.25581399 0.65199131]

[ 1.39236379 0.37256798 0.20336303]]

[[ 0.96462417 0.34291974 0.24251089]

[ 1.05785966 1.65749764 0.82108968]]