Skip to main content

tune.searcher.online_searcher

BaseSearcher Objects

class BaseSearcher()

Abstract class for an online searcher.

ChampionFrontierSearcher Objects

class ChampionFrontierSearcher(BaseSearcher)

The ChampionFrontierSearcher class.

NOTE about the correspondence about this code and the research paper: ChaCha for Online AutoML. This class serves the role of ConfigOralce as described in the paper.

__init__

def __init__(init_config: Dict, space: Optional[Dict] = None, metric: Optional[str] = None, mode: Optional[str] = None, random_seed: Optional[int] = 2345, online_trial_args: Optional[Dict] = {}, nonpoly_searcher_name: Optional[str] = "CFO")

Constructor.

Arguments:

  • init_config - A dictionary of initial configuration.
  • space - A dictionary to specify the search space.
  • metric - A string of the metric name to optimize for.
  • mode - A string in ['min', 'max'] to specify the objective as minimization or maximization.
  • random_seed - An integer of the random seed.
  • online_trial_args - A dictionary to specify the online trial arguments for experimental purpose.
  • nonpoly_searcher_name - A string to specify the search algorithm for nonpoly hyperparameters.

set_search_properties

def set_search_properties(metric: Optional[str] = None, mode: Optional[str] = None, config: Optional[Dict] = {}, setting: Optional[Dict] = {}, init_call: Optional[bool] = False)

Construct search space with the given config, and setup the search.

next_trial

def next_trial()

Return a trial from the _challenger_list.