tf.compat.v1.train.Supervisor

A training helper that checkpoints models and computes summaries.

This class is deprecated. Please use tf.compat.v1.train.MonitoredTrainingSession instead.

The Supervisor is a small wrapper around a Coordinator, a Saver, and a SessionManager that takes care of common needs of TensorFlow training programs.

Use for a single program

with tf.Graph().as_default():
  ...add operations to the graph...
  # Create a Supervisor that will checkpoint the model in '/tmp/mydir'.
  sv = Supervisor(logdir='/tmp/mydir')
  # Get a TensorFlow session managed by the supervisor.
  with sv.managed_session(FLAGS.master) as sess:
    # Use the session to train the graph.
    while not sv.should_stop():
      sess.run(<my_train_op>)

Within the with sv.managed_session() block all variables in the graph have been initialized. In addition, a few services have been started to checkpoint the model and add summaries to the event log.

If the program crashes and is restarted, the managed session automatically reinitialize variables from the most recent checkpoint.

The supervisor is notified of any exception raised by one of the services. After an exception is raised, should_stop() returns True. In that case the training loop should also stop. This is why the training loop has to check for sv.should_stop().

Exceptions that indicate that the training inputs have been exhausted, tf.errors.OutOfRangeError, also cause sv.should_stop() to return True but are not re-raised from the with block: they indicate a normal termination.

Use for multiple replicas

To train with replicas you deploy the same program in a Cluster. One of the tasks must be identified as the chief: the task that handles initialization, checkpoints, summaries, and recovery. The other tasks depend on the chief for these services.

The only change you have to do to the single program code is to indicate if the program is running as the chief.

# Choose a task as the chief. This could be based on server_def.task_index,
# or job_def.name, or job_def.tasks. It's entirely up to the end user.
# But there can be only one *chief*.
is_chief = (server_def.task_index == 0)
server = tf.distribute.Server(server_def)

with tf.Graph().as_default():
  ...add operations to the graph...
  # Create a Supervisor that uses log directory on a shared file system.
  # Indicate if you are the 'chief'
  sv = Supervisor(logdir='/shared_directory/...', is_chief=is_chief)
  # Get a Session in a TensorFlow server on the cluster.
  with sv.managed_session(server.target) as sess:
    # Use the session to train the graph.
    while not sv.should_stop():
      sess.run(<my_train_op>)

In the chief task, the Supervisor works exactly as in the first example above. In the other tasks sv.managed_session() waits for the Model to have been initialized before returning a session to the training code. The non-chief tasks depend on the chief task for initializing the model.

If one of the tasks crashes and restarts, managed_session() checks if the Model is initialized. If yes, it just creates a session and returns it to the training code that proceeds normally. If the model needs to be initialized, the chief task takes care of reinitializing it; the other tasks just wait for the model to have been initialized.

What master string to use

Whether you are running on your machine or in the cluster you can use the following values for the --master flag:

  • Specifying '' requests an in-process session that does not use RPC.

  • Specifying 'local' requests a session that uses the RPC-based "Master interface" to run TensorFlow programs. See tf.train.Server.create_local_server for details.

  • Specifying 'grpc://hostname:port' requests a session that uses the RPC interface to a specific host, and also allows the in-process master to access remote tensorflow workers. Often, it is appropriate to pass server.target (for some tf.distribute.Server named `server).

Advanced use

Launching additional services

managed_session() launches the Checkpoint and Summary services (threads). If you need more services to run you can simply launch them in the block controlled by managed_session().

Example: Start a thread to print losses. We want this thread to run every 60 seconds, so we launch it with sv.loop().

...
sv = Supervisor(logdir='/tmp/mydir')
with sv.managed_session(FLAGS.master) as sess:
  sv.loop(60, print_loss, (sess, ))
  while not sv.should_stop():
    sess.run(my_train_op)
Launching fewer services

managed_session() launches the "summary" and "checkpoint" threads which use either the optionally summary_op and saver passed to the constructor, or default ones created automatically by the supervisor. If you want to run your own summary and checkpointing logic, disable these services by passing None to the summary_op and saver parameters.

Example: Create summaries manually every 100 steps in the chief.

# Create a Supervisor with no automatic summaries.
sv = Supervisor(logdir='/tmp/mydir', is_chief=is_chief, summary_op=None)
# As summary_op was None, managed_session() does not start the
# summary thread.
with sv.managed_session(FLAGS.master) as sess:
  for step in xrange(1000000):
    if sv.should_stop():
      break
    if is_chief and step % 100 == 0:
      # Create the summary every 100 chief steps.
      sv.summary_computed(sess, sess.run(my_summary_op))
    else:
      # Train normally
      sess.run(my_train_op)
Custom model initialization

managed_session() only supports initializing the model by running an init_op or restoring from the latest checkpoint. If you have special initialization needs, see how to specify a local_init_op when creating the supervisor. You can also use the SessionManager directly to create a session and check if it could be initialized automatically.

graph A Graph. The graph that the model will use. Defaults to the default Graph. The supervisor may add operations to the graph before creating a session, but the graph should not be modified by the caller after passing it to the supervisor.
ready_op 1-D string Tensor. This tensor is evaluated by supervisors in prepare_or_wait_for_session() to check if the model is ready to use. The model is considered ready if it r