# 下面是定義一個(gè)卷積層的通用方式
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_intializer(0.0))
  conv = tf.nn.conv2d(input, weights,
    strides=[1, 1, 1, 1], padding='SAME')
  return tf.nn.relu(conv + biases)


# 定義一個(gè)圖片過濾器
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, 32, 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])


# 實(shí)驗(yàn)一：調(diào)用 my_image_filter() 兩次
result1 = my_image_filter(image1)
result2 = my_image_filter(image2)
>>> Raises ValueError(... conv1/weights already exists ...), tf.get_variable()會(huì)檢測已經(jīng)存在的變量是否已經(jīng)共享


# 解決方法一, 可以在設(shè)計(jì)網(wǎng)絡(luò)時(shí)加上一個(gè)布爾型的 reuse 參數(shù) 
with tf.variable_scope("image_filters"):
  result1 = my_image_filter(image1)
with tf.variable_scope("image_filters", reuse=True):
  result2 = my_image_filter(image2)


# 解決方法二
with tf.variable_scope("image_filters") as scope:
  # 下面我們兩次調(diào)用 my_image_filter 函數(shù)，但是由于引入了變量共享機(jī)制
  # 可以看到我們只是創(chuàng)建了一遍網(wǎng)絡(luò)結(jié)構(gòu)。
  result1 = my_image_filter(image1)
  scope.reuse_variables()
  result2 = my_image_filter(image2)


# 解決方法三
with tf.variable_scope("image_filters") as scope:
  result1 = my_image_filter(image1)
with tf.variable_scope(scope, reuse=True):
  result2 = my_image_filter(image2)


# 打印出所有的可訓(xùn)練參變量
vs = tf.trainable_variables()
print('There are %d trainable_variables in the Graph: ' % len(vs))
for v in vs:
  print(v)


# 輸出結(jié)果證明確實(shí)：參變量共享，因?yàn)橹挥兴膫€(gè)變量，沒有創(chuàng)建新的變量。
There are 4 trainable_variables in the Graph: 
Tensor("image_filters/conv1/weights/read:0", shape=(5, 5, 32, 32), dtype=float32)
Tensor("image_filters/conv1/biases/read:0", shape=(32,), dtype=float32)
Tensor("image_filters/conv2/weights/read:0", shape=(5, 5, 32, 32), dtype=float32)
Tensor("image_filters/conv2/biases/read:0", shape=(32,), dtype=float32)

# Returns all variables created with trainable=True in a var_list
var_list = tf.trainable_variables()

init = tf.global_variables_initializer()
sess.run(init)

for var in var_list:
  sess.run(var)