src.baseline.bbo.pso

Module Contents

Classes

PSO

Introduction Particle Swarm Optimization (PSO) optimizer implementation for black-box optimization problems. This class inherits from Basic_Optimizer and utilizes the DEAP library to perform PSO, maintaining a population of particles that iteratively update their positions and velocities to search for the global optimum.

API

class src.baseline.bbo.pso.PSO(config)

Bases: src.environment.optimizer.basic_optimizer.Basic_Optimizer

Introduction

Particle Swarm Optimization (PSO) optimizer implementation for black-box optimization problems.
This class inherits from Basic_Optimizer and utilizes the DEAP library to perform PSO, maintaining a population of particles that iteratively update their positions and velocities to search for the global optimum.

Initialization

Introduction

Initializes the PSO (Particle Swarm Optimization) optimizer with the config object constructed in config.py, setting default hyperparameters and preparing internal state.

Args:

• config (object): Configuration object containing PSO parameters and metadata.

  • The Attributes needed for the PSO are the following:

    • log_interval (int): Interval at which logs are recorded.

    • n_logpoint (int): Number of log points to record. Default is 50.

    • full_meta_data (bool): Flag indicating whether to use full meta data. Default is False.

    • maxFEs (int): Maximum number of function evaluations allowed. Default value depends on the type of the problem.

    • phi1 (float): Cognitive coefficient for particle velocity update. Default is 2.0.

    • phi2 (float): Social coefficient for particle velocity update. Default is 2.0.

    • population_size (int): Size of the particle population. Default is 50.

Attributes:

• __config (object): Stores the configuration object.

• __toolbox (object or None): Placeholder for the optimization toolbox, initialized as None.

• __creator (object or None): Placeholder for the creator utility, initialized as None.

Notes:

• Sets default values for phi1, phi2, and population_size in the configuration.

__str__()

Returns a string representation of the PSO (Particle Swarm Optimization) class.

Returns:

str: The string "PSO", representing the class name.

run_episode(problem)

Introduction

Executes a single episode of Particle Swarm Optimization (PSO) on the given optimization problem, tracking the best solution found and optionally collecting meta-data about the optimization process.

Args:

• problem (object): An object representing the optimization problem to solve. Must have attributes lb (lower bounds), ub (upper bounds), dim (dimension), eval (evaluation function), and optionally optimum (known optimum value).

Returns:

• dict: A dictionary containing:

  • ‘cost’ (list of float): The best fitness value found at each logging interval.

  • ‘fes’ (int): The total number of function evaluations performed.

  • ‘metadata’ (dict, optional): If self.full_meta_data is True, includes:

    • ‘X’ (list of np.ndarray): The population positions at each logging interval.

    • ‘Cost’ (list of float): The fitness values of the population at each logging interval.

Raises:

• AttributeError: If required attributes are missing from the problem object.

• Exception: For errors during the optimization process, such as invalid configuration or evaluation failures.