Tune Execution (tune.Tuner)#

Tuner#

Tuner([trainable, param_space, tune_config, ...])

Tuner is the recommended way of launching hyperparameter tuning jobs with Ray Tune.

Tuner.fit()

Executes hyperparameter tuning job as configured and returns result.

Tuner.get_results()

Get results of a hyperparameter tuning run.

Tuner Configuration#

TuneConfig([mode, metric, search_alg, ...])

Tune specific configs.

See also

The Tuner constructor also takes in a air.RunConfig.

Restoring a Tuner#

Tuner.restore(path, trainable[, ...])

Restores Tuner after a previously failed run.

Tuner.can_restore(path[, storage_filesystem])

Checks whether a given directory contains a restorable Tune experiment.

tune.run_experiments#

run_experiments(experiments[, scheduler, ...])

Runs and blocks until all trials finish.

Experiment(name, run, *[, stop, ...])

Tracks experiment specifications.