Python training_util.get_global_step方法代码示例

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def assert_or_get_global_step(graph=None, global_step_tensor=None):
  """Verifies that a global step tensor is valid or gets one if None is given.

  If `global_step_tensor` is not None, check that it is a valid global step
  tensor (using `assert_global_step`). Otherwise find a global step tensor using
  `get_global_step` and return it.

  Args:
    graph: The graph to find the global step tensor for.
    global_step_tensor: The tensor to check for suitability as a global step. If
      None is given (the default), find a global step tensor.

  Returns:
    A tensor suitable as a global step, or `None` if none was provided and none
    was found.
  """
  if global_step_tensor is None:
    # Get the global step tensor the same way the supervisor would.
    global_step_tensor = get_global_step(graph)
  else:
    assert_global_step(global_step_tensor)
  return global_step_tensor 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def assert_or_get_global_step(graph=None, global_step_tensor=None):
  """Verifies that a global step tensor is valid or gets one if None is given.

  If `global_step_tensor` is not None, check that it is a valid global step
  tensor (using `assert_global_step`). Otherwise find a global step tensor using
  `get_global_step` and return it.

  Args:
    graph: The graph to find the global step tensor for.
    global_step_tensor: The tensor to check for suitability as a global step.
      If None is given (the default), find a global step tensor.

  Returns:
    A tensor suitable as a global step, or `None` if none was provided and none
    was found.
  """
  if global_step_tensor is None:
    # Get the global step tensor the same way the supervisor would.
    global_step_tensor = get_global_step(graph)
  else:
    assert_global_step(global_step_tensor)
  return global_step_tensor 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def create_global_step(graph=None):
  """Create global step tensor in graph.

  Args:
    graph: The graph in which to create the global step. If missing, use default
        graph.

  Returns:
    Global step tensor.

  Raises:
    ValueError: if global step key is already defined.
  """
  graph = ops.get_default_graph() if graph is None else graph
  if get_global_step(graph) is not None:
    raise ValueError('"global_step" already exists.')
  # Create in proper graph and base name_scope.
  with graph.as_default() as g, g.name_scope(None):
    collections = [ops.GraphKeys.GLOBAL_VARIABLES, ops.GraphKeys.GLOBAL_STEP]
    return variable(ops.GraphKeys.GLOBAL_STEP, shape=[], dtype=dtypes.int64,
                    initializer=init_ops.zeros_initializer, trainable=False,
                    collections=collections) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def begin(self):
    self._global_step_tensor = training_util.get_global_step()
    self._stop_var = self._get_or_create_stop_var_with_aggregation()
    assert distribution_strategy_context.in_cross_replica_context()

    strategy = distribution_strategy_context.get_strategy()
    self._stop_placeholder = None

    def stop_op_fn(var):
      placeholder = array_ops.placeholder_with_default(
          0, tuple(), name='stop_value')
      if self._stop_placeholder is None:
        self._stop_placeholder = placeholder
      return var.assign_add(placeholder)

    self._stop_op = strategy.run(
        stop_op_fn, args=(self._stop_var,)) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def get_global_step(graph=None):
  return training_util.get_global_step(graph) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def begin(self):
    self._global_step_tensor = training_util.get_global_step()
    if self._global_step_tensor is None:
      raise RuntimeError("Global step should be created to use StopAtStepHook.") 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def end(self, session):
    last_step = session.run(training_util.get_global_step())
    if last_step != self._timer.last_triggered_step():
      self._save(last_step, session)
    for l in self._listeners:
      l.end(session, last_step) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def begin(self):
    self._next_step = None
    self._global_step_tensor = training_util.get_global_step()
    if self._global_step_tensor is None:
      raise RuntimeError(
          "Global step should be created to use ProfilerHook.") 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def compute_gradients(self, loss, *args, **kwargs):
    # Record current global step for worker.
    with ops.colocate_with(loss):
      self._local_step = training_util.get_global_step() + 0

    with ops.control_dependencies([self._local_step]):
      loss = gen_array_ops.identity(loss)
      return self._opt.compute_gradients(loss, *args, **kwargs) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def create_global_step(graph=None):
  """Create global step tensor in graph.

  Args:
    graph: The graph in which to create the global step. If missing, use default
        graph.

  Returns:
    Global step tensor.

  Raises:
    ValueError: if global step key is already defined.
  """
  graph = ops.get_default_graph() if graph is None else graph
  if get_global_step(graph) is not None:
    raise ValueError('"global_step" already exists.')
  # Create in proper graph and base name_scope.
  with graph.as_default() as g, g.name_scope(None):
    collections = [ops.GraphKeys.GLOBAL_VARIABLES, ops.GraphKeys.GLOBAL_STEP]
    return variable(
        ops.GraphKeys.GLOBAL_STEP,
        shape=[],
        dtype=dtypes.int64,
        initializer=init_ops.zeros_initializer(),
        trainable=False,
        collections=collections) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def get_or_create_global_step(graph=None):
  """Returns and create (if necessary) the global step variable.

  Args:
    graph: The graph in which to create the global step. If missing, use default
        graph.

  Returns:
    the tensor representing the global step variable.
  """
  graph = ops.get_default_graph() if graph is None else graph
  globalstep = get_global_step(graph)
  if globalstep is None:
    globalstep = create_global_step(graph)
  return globalstep 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def begin(self):
    self._global_step_tensor = training_util.get_global_step()
    if self._global_step_tensor is None:
      raise RuntimeError("Global step should be created to use UpdateGlobalStepHook.")
    ops.get_default_graph()._unsafe_unfinalize()
    self._updated_global_step = state_ops.assign_add(self._global_step_tensor, 1, use_locking=True) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def _create_global_step(graph):
  graph = graph or ops.get_default_graph()
  if training.get_global_step(graph) is not None:
    raise ValueError('"global_step" already exists.')
  # Create in proper graph and base name_scope.
  with graph.as_default() as g, g.name_scope(None):
    return variable_scope.get_variable(
        ops.GraphKeys.GLOBAL_STEP,
        shape=[],
        dtype=dtypes.int64,
        initializer=init_ops.zeros_initializer(),
        trainable=False,
        use_resource=True,
        collections=[ops.GraphKeys.GLOBAL_VARIABLES, ops.GraphKeys.GLOBAL_STEP]) 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def begin(self):
    self._global_step_tensor = training_util.get_global_step()
    if self._global_step_tensor is None:
      raise RuntimeError('Global step should be created.')

    self._iterations_per_loop_var = _create_or_get_iterations_per_loop() 

# 需要导入模块: from tensorflow.python.training import training_util [as 别名]
# 或者: from tensorflow.python.training.training_util import get_global_step [as 别名]
def begin(self):
    self._global_step_tensor = training_util.get_global_step()
    if self._global_step_tensor is None:
      raise RuntimeError(
          'Global step should be created to use StepCounterHook.')