Python tf.Graph.control_dependencies用法及代码示例

用法

control_dependencies(
    control_inputs
)

返回指定控制依赖关系的上下文管理器。

与 with 关键字一起使用以指定在上下文中构造的所有操作都应具有对 control_inputs 的控制依赖项。例如：

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

可以嵌套对control_dependencies() 的多次调用，在这种情况下，新的Operation 将对来自所有活动上下文的control_inputs 的联合具有控制依赖关系。

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`.

你可以通过 None 来清除控制依赖：

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：控制依赖上下文适用只要到在上下文中构建的操作。仅在上下文中使用操作或张量不会添加控制依赖项。以下示例说明了这一点：

# 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)

另请注意，尽管在此范围内创建的操作的执行将触发依赖项的执行，但在此范围内创建的操作可能仍会从正常的 tensorflow 图中删除。例如，在以下代码片段中，从不执行依赖项：

loss = model.loss()
  with tf.control_dependencies(dependencies):
    loss = loss + tf.constant(1)  # note:dependencies ignored in the
                                  # backward pass
  return tf.gradients(loss, model.variables)

这是因为评估梯度图不需要评估前向传递中创建的 constant(1) op。