Configuration Repository

ConfigItem(config_id, config, results) dataclass

Data class to store information regarding a specific configuration.

The results for this configuration are stored in the results dict, using the fidelity it has been evaluated on as keys.

ResultItem(score, cost, info) dataclass

Data class storing the result information of a specific configuration + fidelity.

ConfigRepository()

Bookkeeps all configurations used throughout the course of the optimization.

Keeps track of the configurations and their results on the different fidelitites. A new configuration is announced via announce_config. After evaluating the configuration on the specified fidelity, use tell_result to log the achieved performance, cost etc.

The configurations are stored in a list of ConfigItem.

Initializes the class by calling self.reset.

Source code in src/dehb/utils/config_repository.py

def __init__(self) -> None:
    """Initializes the class by calling `self.reset`."""
    self.configs : list[ConfigItem]
    self.reset()

reset()

Resets the config repository, clearing all collected configurations and results.

Source code in src/dehb/utils/config_repository.py

def reset(self) -> None:
    """Resets the config repository, clearing all collected configurations and results."""
    self.configs = []

announce_config(config, fidelity=None)

Announces a new configuration with the respective fidelity it should be evaluated on.

The configuration is then added to the list of so far seen configurations and the ID of the configuration is returned.

PARAMETER	DESCRIPTION
config	New configuration TYPE: ndarray
fidelity	Fidelity on which config is evaluated or None. Defaults to None. TYPE: float DEFAULT: None

RETURNS	DESCRIPTION
int	ID of configuration TYPE: int

Source code in src/dehb/utils/config_repository.py

def announce_config(self, config: np.ndarray, fidelity=None) -> int:
    """Announces a new configuration with the respective fidelity it should be evaluated on.

    The configuration is then added to the list of so far seen configurations and the ID of the
    configuration is returned.

    Args:
        config (np.ndarray): New configuration
        fidelity (float, optional): Fidelity on which `config` is evaluated or None.
                                    Defaults to None.

    Returns:
        int: ID of configuration
    """
    config_id = len(self.configs)
    fidelity = float(fidelity or 0)
    result_dict = {
            fidelity: ResultItem(np.inf, -1, {}),
        }
    config_item = ConfigItem(config_id, config.copy(), result_dict)
    self.configs.append(config_item)
    return config_id

announce_population(population, fidelity=None)

Announce population, retrieving ids for the population.

PARAMETER	DESCRIPTION
population	Population to announce TYPE: ndarray
fidelity	Fidelity on which pop is evaluated or None. Defaults to None. TYPE: float DEFAULT: None

RETURNS	DESCRIPTION
ndarray	np.ndarray: population ids

Source code in src/dehb/utils/config_repository.py

def announce_population(self, population: np.ndarray, fidelity=None) -> np.ndarray:
    """Announce population, retrieving ids for the population.

    Args:
        population (np.ndarray): Population to announce
        fidelity (float, optional): Fidelity on which pop is evaluated or None.
                                    Defaults to None.

    Returns:
        np.ndarray: population ids
    """
    population_ids = []
    for indiv in population:
        conf_id = self.announce_config(indiv, float(fidelity or 0))
        population_ids.append(conf_id)
    return np.array(population_ids)

announce_fidelity(config_id, fidelity)

Announce the evaluation of a new fidelity for a given config.

This function may only be used if the config already exists in the repository. Note: This function is currently unused, but might be used later in order to allow for continuation.

PARAMETER	DESCRIPTION
config_id	ID of Configuration TYPE: int
fidelity	Fidelity on which the config will be evaluated TYPE: float

Source code in src/dehb/utils/config_repository.py

def announce_fidelity(self, config_id: int, fidelity: float):
    """Announce the evaluation of a new fidelity for a given config.

    This function may only be used if the config already exists in the repository.
    Note: This function is currently unused, but might be used later in order to
    allow for continuation.

    Args:
        config_id (int): ID of Configuration
        fidelity (float): Fidelity on which the config will be evaluated
    """
    try:
        config_item = self.configs[config_id]
    except IndexError as e:
        raise IndexError("Config with the given ID can not be found.") from e

    result_item = {
            fidelity: ResultItem(np.inf, -1, {}),
        }
    config_item.results[fidelity] = result_item

tell_result(config_id, fidelity, score, cost, info)

Logs the achieved performance, cost etc. of a specific configuration-fidelity pair.

PARAMETER	DESCRIPTION
config_id	ID of evaluated configuration TYPE: int
fidelity	Fidelity on which configuration has been evaluated. TYPE: float
score	Achieved score, given by objective function TYPE: float
cost	Cost, given by objective function TYPE: float
info	Run info, given by objective function TYPE: dict

Source code in src/dehb/utils/config_repository.py

def tell_result(self, config_id: int, fidelity: float, score: float, cost: float, info: dict):
    """Logs the achieved performance, cost etc. of a specific configuration-fidelity pair.

    Args:
        config_id (int): ID of evaluated configuration
        fidelity (float): Fidelity on which configuration has been evaluated.
        score (float): Achieved score, given by objective function
        cost (float): Cost, given by objective function
        info (dict): Run info, given by objective function
    """
    try:
        config_item = self.configs[config_id]
    except IndexError as e:
        raise IndexError("Config with the given ID can not be found.") from e

    # If configuration has been promoted, there is no fidelity information yet
    if fidelity not in config_item.results:
        config_item.results[fidelity] = ResultItem(score, cost, info)
    else:
        # ResultItem already given for specified fidelity --> update entries
        config_item.results[fidelity].score = score
        config_item.results[fidelity].cost = cost
        config_item.results[fidelity].info = info

get(config_id)

Get the configuration with the given ID.

PARAMETER	DESCRIPTION
config_id	ID of config TYPE: int

RETURNS	DESCRIPTION
ndarray	np.ndarray: Config in hypercube representation

Source code in src/dehb/utils/config_repository.py

def get(self, config_id: int) -> np.ndarray:
    """Get the configuration with the given ID.

    Args:
        config_id (int): ID of config

    Returns:
        np.ndarray: Config in hypercube representation
    """
    try:
        config_item = self.configs[config_id]
    except IndexError as e:
        raise IndexError("Config with the given ID can not be found.") from e
    return config_item.config

save_state(save_path)

Saves the current state to save_path.

PARAMETER	DESCRIPTION
save_path	Path where the state should be saved to. TYPE: Path

Source code in src/dehb/utils/config_repository.py

def save_state(self, save_path: Path):
    """Saves the current state to `save_path`.

    Args:
        save_path (Path): Path where the state should be saved to.
    """
    with save_path.open("w") as f:
        serialized_data = []
        for config in self.configs:
            serialized_config = asdict(config)
            serialized_config["config"] = serialized_config["config"].tolist()
            serialized_data.append(serialized_config)
        json.dump(serialized_data, f, indent=2)