tfdbg.GrpcDebugHook

Class GrpcDebugHook

Inherits From: SessionRunHook

Defined in tensorflow/python/debug/wrappers/hooks.py.

A hook that streams debugger-related events to any grpc_debug_server.

For example, the debugger data server is a grpc_debug_server. The debugger data server writes debugger-related events it receives via GRPC to logdir. This enables debugging features in Tensorboard such as health pills.

When the arguments of debug_utils.watch_graph changes, strongly consider changing arguments here too so that features are available to tflearn users.

Can be used as a monitor/hook for tf.train.MonitoredSessions and tf.contrib.learn's Estimators and Experiments.

Methods

__init__

__init__(
    grpc_debug_server_addresses,
    watch_fn=None,
    thread_name_filter=None,
    log_usage=True
)

Constructs a GrpcDebugHook.

Args:

  • grpc_debug_server_addresses: (list of str) A list of the gRPC debug server addresses, in the format of host:port, with or without the "grpc://" prefix. For example: ["localhost:7000", "192.168.0.2:8000"]
  • watch_fn: A function that allows for customizing which ops to watch at which specific steps. See doc of dumping_wrapper.DumpingDebugWrapperSession.__init__ for details.
  • thread_name_filter: Regular-expression white list for threads on which the wrapper session will be active. See doc of BaseDebugWrapperSession for more details.
  • log_usage: (bool) Whether usage is to be logged.

after_create_session

after_create_session(
    session,
    coord
)

Called when new TensorFlow session is created.

This is called to signal the hooks that a new session has been created. This has two essential differences with the situation in which begin is called:

  • When this is called, the graph is finalized and ops can no longer be added to the graph.
  • This method will also be called as a result of recovering a wrapped session, not only at the beginning of the overall session.

Args:

  • session: A TensorFlow Session that has been created.
  • coord: A Coordinator object which keeps track of all threads.

after_run

after_run(
    run_context,
    run_values
)

Called after each call to run().

The run_values argument contains results of requested ops/tensors by before_run().

The run_context argument is the same one send to before_run call. run_context.request_stop() can be called to stop the iteration.

If session.run() raises any exceptions then after_run() is not called.

Args:

  • run_context: A SessionRunContext object.
  • run_values: A SessionRunValues object.

before_run

before_run(run_context)

Called right before a session is run.

Args:

  • run_context: A session_run_hook.SessionRunContext. Encapsulates information on the run.

Returns:

A session_run_hook.SessionRunArgs object.

begin

begin()

Called once before using the session.

When called, the default graph is the one that will be launched in the session. The hook can modify the graph by adding new operations to it. After the begin() call the graph will be finalized and the other callbacks can not modify the graph anymore. Second call of begin() on the same graph, should not change the graph.

end

end(session)

Called at the end of session.

The session argument can be used in case the hook wants to run final ops, such as saving a last checkpoint.

If session.run() raises exception other than OutOfRangeError or StopIteration then end() is not called. Note the difference between end() and after_run() behavior when session.run() raises OutOfRangeError or StopIteration. In that case end() is called but after_run() is not called.

Args:

  • session: A TensorFlow Session that will be soon closed.