ray.tune.search.Searcher#

class ray.tune.search.Searcher(metric: str | None = None, mode: str | None = None)[source]#

Abstract class for wrapping suggesting algorithms.

Custom algorithms can extend this class easily by overriding the suggest method provide generated parameters for the trials.

Any subclass that implements __init__ must also call the constructor of this class: super(Subclass, self).__init__(...).

To track suggestions and their corresponding evaluations, the method suggest will be passed a trial_id, which will be used in subsequent notifications.

Not all implementations support multi objectives.

Note to Tune developers: If a new searcher is added, please update air/_internal/usage.py.

Parameters:

metric – The training result objective value attribute. If list then list of training result objective value attributes
mode – If string One of {min, max}. If list then list of max and min, determines whether objective is minimizing or maximizing the metric attribute. Must match type of metric.

class ExampleSearch(Searcher):
    def __init__(self, metric="mean_loss", mode="min", **kwargs):
        super(ExampleSearch, self).__init__(
            metric=metric, mode=mode, **kwargs)
        self.optimizer = Optimizer()
        self.configurations = {}

    def suggest(self, trial_id):
        configuration = self.optimizer.query()
        self.configurations[trial_id] = configuration

    def on_trial_complete(self, trial_id, result, **kwargs):
        configuration = self.configurations[trial_id]
        if result and self.metric in result:
            self.optimizer.update(configuration, result[self.metric])

tuner = tune.Tuner(
    trainable_function,
    tune_config=tune.TuneConfig(
        search_alg=ExampleSearch()
    )
)
tuner.fit()

DeveloperAPI: This API may change across minor Ray releases.

Methods

`add_evaluated_point`	Pass results from a point that has been evaluated separately.
`add_evaluated_trials`	Pass results from trials that have been evaluated separately.
`on_trial_complete`	Notification for the completion of trial.
`on_trial_result`	Optional notification for result during training.
`restore`	Restore state for this search algorithm
`restore_from_dir`	Restores the state of a searcher from a given checkpoint_dir.
`save`	Save state to path for this search algorithm.
`save_to_dir`	Automatically saves the given searcher to the checkpoint_dir.
`set_max_concurrency`	Set max concurrent trials this searcher can run.
`set_search_properties`	Pass search properties to searcher.
`suggest`	Queries the algorithm to retrieve the next set of parameters.

Attributes

`CKPT_FILE_TMPL`
`FINISHED`
`metric`	The training result objective value attribute.
`mode`	Specifies if minimizing or maximizing the metric.