ray.tune.logger.TBXLoggerCallback#

class ray.tune.logger.TBXLoggerCallback[source]#

Bases: ray.tune.logger.logger.LoggerCallback

TensorBoardX Logger.

Note that hparams will be written only after a trial has terminated. This logger automatically flattens nested dicts to show on TensorBoard:

{“a”: {“b”: 1, “c”: 2}} -> {“a/b”: 1, “a/c”: 2}

PublicAPI: This API is stable across Ray releases.

Methods

get_state()

Get the state of the callback.

log_trial_restore(trial)

Handle logging when a trial restores.

log_trial_save(trial)

Handle logging when a trial saves a checkpoint.

on_checkpoint(iteration, trials, trial, ...)

Called after a trial saved a checkpoint with Tune.

on_experiment_end(trials, **info)

Called after experiment is over and all trials have concluded.

on_step_begin(iteration, trials, **info)

Called at the start of each tuning loop step.

on_step_end(iteration, trials, **info)

Called at the end of each tuning loop step.

on_trial_recover(iteration, trials, trial, ...)

Called after a trial instance failed (errored) but the trial is scheduled for retry.

set_state(state)

Set the state of the callback.

setup([stop, num_samples, total_num_samples])

Called once at the very beginning of training.

Attributes

VALID_HPARAMS

VALID_NP_HPARAMS