tf.random.truncated_normal()-截断正态分布

tf.random.truncated_normal()-截断正态分布

一、总结

一句话总结：

sigmoid激活函数，用截断的正态分布更好，因为这样就不会有两侧的梯度消失的情况

 

 

二、tf.random.truncated

转自或参考：tf.random.truncated
https://blog.csdn.net/qq_39894692/article/details/101635922

来自https://tensorflow.google.cn/versions/r1.15/api_docs/python/tf/random/truncated_normal?hl=en

Outputs random values from a truncated normal distribution.

此函数别名有：

• tf.compat.v1.random.truncated_normal
• tf.compat.v1.truncated_normal
• tf.compat.v2.random.truncated_normal
• tf.truncated_normal

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

Args:

• shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
• mean: A 0-D Tensor or Python value of type dtype. The mean of the truncated normal distribution.均值默认为0
• stddev: A 0-D Tensor or Python value of type dtype. The standard deviation of the normal distribution, before truncation.截断前正态分布的标准偏差，默认为1.0
• dtype: The type of the output.
• seed: A Python integer. Used to create a random seed for the distribution. Seetf.compat.v1.set_random_seed for behavior.
• name: A name for the operation (optional).