tf.control_dependencies()作用及用法

参考的博客：https://blog.csdn.net/PKU_Jade/article/details/73498753
在facenet .py的line 75-77 有引用该函数。

with tf.control_dependencies([centers]): # 操作依赖，loss内的操作，将会再centers后再进行。
        loss = tf.reduce_mean(tf.square(features - centers_batch)) # 
    return loss, centers

具体的用法如下。

with g.control_dependencies([a, b, c]):
  # `d` and `e` will only run after `a`, `b`, and `c` have executed.
  d = ...
  e = ...

可以嵌套control_dependencies 使用

with g.control_dependencies([a, b]):
  # Ops constructed here run after `a` and `b`.
  with g.control_dependencies([c, d]):
    # Ops constructed here run after `a`, `b`, `c`, and `d`

.

可以传入None 来消除依赖：

with g.control_dependencies([a, b]):
  # Ops constructed here run after `a` and `b`.
  with g.control_dependencies(None):
    # Ops constructed here run normally, not waiting for either `a` or `b`.
    with g.control_dependencies([c, d]):
      # Ops constructed here run after `c` and `d`, also not waiting
      # for either `a` or `b`.

注意：
控制依赖只对那些在上下文环境中建立的操作有效，仅仅在context中使用一个操作或张量是没用的

# WRONG
def my_func(pred, tensor):
  t = tf.matmul(tensor, tensor)
  with tf.control_dependencies([pred]):
    # The matmul op is created outside the context, so no control
    # dependency will be added.
    return t

# RIGHT
def my_func(pred, tensor):
  with tf.control_dependencies([pred]):
    # The matmul op is created in the context, so a control dependency
    # will be added.
    return tf.matmul(tensor, tensor)

例子：
在训练模型时我们每步训练可能要执行两种操作，op a, b 这时我们就可以使用如下代码：

with tf.control_dependencies([a, b]):
    c= tf.no_op(name='train')#tf.no_op；什么也不做
sess.run(c)

在这样简单的要求下，可以将上面代码替换为：

c= tf.group([a, b])
sess.run(c)