TensorFlow学习笔记4——变量共享

因为最近在研究生成对抗网络GAN,在读别人的代码时发现了 with tf.variable_scope(self.name_scope_conv, reuse = reuse): 这样一条语句,查阅官方文档时明白了这是TensorFlow的变量共享机制。

举个例子:当我们研究生成对抗网络GAN的时候,判别器的任务是,如果接收到的是生成器生成的图像,判别器就尝试优化自己的网络结构来使自己输出0,如果接收到的是来自真实数据的图像,那么就尝试优化自己的网络结构来使自己输出1。也就是说,生成图像和真实图像经过判别器的时候,要共享同一套变量,所以TensorFlow引入了变量共享机制。

变量共享主要涉及到两个函数:  tf.get_variable(<name>, <shape>, <initializer>)  和  tf.variable_scope(<scope_name>)  。

1. tf.get_variable(<name>, <shape>, <initializer>)

例如,我们搭建一个卷积层:

def conv_relu(input, kernel_shape, bias_shape):
    # Create variable named "weights".
    weights = tf.get_variable("weights", kernel_shape,
        initializer=tf.random_normal_initializer())
    # Create variable named "biases".
    biases = tf.get_variable("biases", bias_shape,
        initializer=tf.constant_initializer(0.0))
    conv = tf.nn.conv2d(input, weights,
        strides=[1, 1, 1, 1], padding='SAME')
    return tf.nn.relu(conv + biases)

然后,我们调用两次:

input1 = tf.random_normal([1,10,10,32])
input2 = tf.random_normal([1,20,20,32])
x = conv_relu(input1, kernel_shape=[5, 5, 1, 32], bias_shape=[32])
x = conv_relu(x, kernel_shape=[5, 5, 32, 32], bias_shape = [32])  # This fails.

会发现报错信息。因为执行的命令不明确:第二次调用时是创建一套新的变量(weights,biases)还是再次使用已存在的那一套变量(第一次调用时生成的weights和biases)呢?

这时就需要用到第二个函数: tf.variable_scope(<scope_name>)  

2. tf.variable_scope(<scope_name>)

请看例子:

def my_image_filter(input_images):
    with tf.variable_scope("conv1"):
        # Variables created here will be named "conv1/weights", "conv1/biases".
        relu1 = conv_relu(input_images, [5, 5, 1, 32], [32])
    with tf.variable_scope("conv2"):
        # Variables created here will be named "conv2/weights", "conv2/biases".
        return conv_relu(relu1, [5, 5, 32, 32], [32])

在不同的域内会生成不同的变量。

如果想要变量共享,TensorFlow提供了两种方法:

1. 设置  reuse=True 

with tf.variable_scope("model"):
  output1 = my_image_filter(input1)
with tf.variable_scope("model", reuse=True):
  output2 = my_image_filter(input2)

2. 调用 scope.reuse_variables() 

with tf.variable_scope("model") as scope:
  output1 = my_image_filter(input1)
  scope.reuse_variables()
  output2 = my_image_filter(input2)

注:在官方文档的最后有这样一段话:Since depending on exact string names of scopes can feel dangerous, it's also possible to initialize a variable scope based on another one:

with tf.variable_scope("model") as scope:
  output1 = my_image_filter(input1)
with tf.variable_scope(scope, reuse=True):
  output2 = my_image_filter(input2)
原文地址:https://www.cnblogs.com/congyucn/p/7427632.html