deephyper.search.nas.AgEBO

class deephyper.search.nas.AgEBO(problem, evaluator, random_state: Optional[int] = None, log_dir: str = '.', verbose: int = 0, population_size: int = 100, sample_size: int = 10, surrogate_model: str = 'RF', acq_func: str = 'UCB', kappa: float = 0.001, xi: float = 1e-06, n_points: int = 10000, liar_strategy: str = 'cl_max', n_jobs: int = 1, mode: str = 'async', **kwargs)

Bases: deephyper.search.nas._regevo.RegularizedEvolution

Aging evolution with Bayesian Optimization.

This algorithm build on the Regularized Evolution. It cumulates Hyperparameter optimization with Bayesian optimisation and Neural architecture search with regularized evolution.

Parameters

• problem (NaProblem) – Neural architecture search problem describing the search space to explore.

• evaluator (Evaluator) – An Evaluator instance responsible of distributing the tasks.

• random_state (int, optional) – Random seed. Defaults to None.

• log_dir (str, optional) – Log directory where search’s results are saved. Defaults to “.”.

• verbose (int, optional) – Indicate the verbosity level of the search. Defaults to 0.

• population_size (int, optional) – the number of individuals to keep in the population. Defaults to 100.

• sample_size (int, optional) – the number of individuals that should participate in each tournament. Defaults to 10.

• surrogate_model (str, optional) – Surrogate model used by the Bayesian optimization. Can be a value in ["RF", "ET", "GBRT", "DUMMY"]. Defaults to "RF".

• acq_func (str, optional) – Acquisition function used by the Bayesian optimization. Can be a value in ["UCB", "EI", "PI", "gp_hedge"]. Defaults to "UCB".

• kappa (float, optional) – Manage the exploration/exploitation tradeoff for the “UCB” acquisition function. Defaults to 0.001 for strong exploitation.

• xi (float, optional) – Manage the exploration/exploitation tradeoff of "EI" and "PI" acquisition function. Defaults to 0.000001 for strong exploitation.

• n_points (int, optional) – The number of configurations sampled from the search space to infer each batch of new evaluated configurations. Defaults to 10000.

• liar_strategy (str, optional) – Definition of the constant value use for the Liar strategy. Can be a value in ["cl_min", "cl_mean", "cl_max"] . Defaults to "cl_max".

• n_jobs (int, optional) – Number of parallel processes used to fit the surrogate model of the Bayesian optimization. A value of -1 will use all available cores. Defaults to 1.

• mode (str, optional) – Define if the search should be asynchronous or batch synchronous. Choice in [“sync”, “async”]. Defaults to “async”.

Methods

dump_context	Dumps the context in the log folder.
search	Execute the search algorithm.
terminate	Terminate the search.
to_json	Returns a json version of the search object.

dump_context()

Dumps the context in the log folder.

search(max_evals: int = - 1, timeout: Optional[int] = None)

Execute the search algorithm.

Parameters

• max_evals (int, optional) – The maximum number of evaluations of the run function to perform before stopping the search. Defaults to -1, will run indefinitely.

• timeout (int, optional) – The time budget (in seconds) of the search before stopping. Defaults to None, will not impose a time budget.

Returns

a pandas DataFrame containing the evaluations performed or None if the search could not evaluate any configuration.

Return type

DataFrame

terminate()

Terminate the search.

Raises

SearchTerminationError – raised when the search is terminated with SIGALARM

to_json()

Returns a json version of the search object.