在Tensorflow中使用泄漏的relu [英] using leaky relu in Tensorflow

问题描述

如何将G_h1 = tf.nn.relu(tf.matmul(z, G_W1) + G_b1)更改为泄漏的relu?我尝试使用max(value, 0,01*value)遍历张量，但得到TypeError: Using a tf.Tensor as a Python bool is not allowed.

How can I change G_h1 = tf.nn.relu(tf.matmul(z, G_W1) + G_b1) to leaky relu? I have tried looping over the tensor using max(value, 0,01*value) but I get TypeError: Using a tf.Tensor as a Python bool is not allowed.

我还尝试在Tensorflow github上的relu上找到源代码，以便可以将其修改为泄漏的relu，但是我找不到它..

I also tried to find the source code on relu on Tensorflow github so that I can modify it to leaky relu but I couldn't find it..

推荐答案

您可以根据tf.relu编写一个，例如:

You could write one based on tf.relu, something like:

def lrelu(x, alpha):
  return tf.nn.relu(x) - alpha * tf.nn.relu(-x)

编辑

Tensorflow 1.4现在具有本机 tf.nn.leaky_relu .

Tensorflow 1.4 now has a native tf.nn.leaky_relu.

这篇关于在Tensorflow中使用泄漏的relu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！