mealpy.sota_based package

mealpy.sota_based.IMODE module

class mealpy.sota_based.IMODE.OriginalIMODE(epoch: int = 10000, pop_size: int = 100, memory_size: int = 5, archive_size: int = 20, **kwargs: object)

Bases: Optimizer

The original version of: Improved Multi-operator Differential Evolution Algorithm (IMODE)

Links

1. https://doi.org/10.1109/CEC48606.2020.9185577

2. This version is conversion from the original MATLAB code available at CEC competition github.

Examples

>>> import numpy as np
>>> from mealpy import FloatVar, IMODE
>>>
>>> def objective_function(solution):
>>>     return np.sum(solution**2)
>>>
>>> problem_dict = {
>>>     "bounds": FloatVar(lb=(-10.,) * 30, ub=(10.,) * 30, name="delta"),
>>>     "minmax": "min",
>>>     "obj_func": objective_function
>>> }
>>>
>>> model = IMODE.OriginalIMODE(epoch=1000, pop_size=50, memory_size=5, archive_size=20)
>>> g_best = model.solve(problem_dict)
>>> print(f"Solution: {g_best.solution}, Fitness: {g_best.target.fitness}")
>>> print(f"Solution: {model.g_best.solution}, Fitness: {model.g_best.target.fitness}")

References

[1] Sallam, K. M., Elsayed, S. M., Chakrabortty, R. K., & Ryan, M. J. (2020, July). Improved multi-operator differential evolution algorithm for solving unconstrained problems. In 2020 IEEE congress on evolutionary computation (CEC) (pp. 1-8). IEEE.

before_main_loop()

evolve(epoch)

The main operations (equations) of algorithm. Inherit from Optimizer class

Parameters

epoch (int) – The current iteration

initialize_variables()

mealpy.sota_based.LSHADEcnEpSin module

class mealpy.sota_based.LSHADEcnEpSin.OriginalLSHADEcnEpSin(epoch: int = 10000, pop_size: int = 100, miu_f: float = 0.5, miu_cr: float = 0.5, freq: float = 0.5, memory_size: int = 5, ps: float = 0.5, pc: float = 0.4, pop_size_min: int = 10, **kwargs: object)

Bases: Optimizer

The original version of: Ensemble sinusoidal differential covariance matrix adaptation with Euclidean neighborhood (LSHADEcnEpSin)

Links

1. https://doi.org/10.1109/CEC.2017.7969336

Examples

>>> import numpy as np
>>> from mealpy import FloatVar, LSHADEcnEpSin
>>>
>>> def objective_function(solution):
>>>     return np.sum(solution**2)
>>>
>>> problem_dict = {
>>>     "bounds": FloatVar(lb=(-10.,) * 30, ub=(10.,) * 30, name="delta"),
>>>     "minmax": "min",
>>>     "obj_func": objective_function
>>> }
>>>
>>> model = LSHADEcnEpSin.OriginalLSHADEcnEpSin(epoch=1000, pop_size=50, miu_f = 0.5, miu_cr = 0.5,
>>>                                freq = 0.5, memory_size = 5, ps = 0.5, pc = 0.4, pop_size_min = 10)
>>> g_best = model.solve(problem_dict)
>>> print(f"Solution: {g_best.solution}, Fitness: {g_best.target.fitness}")
>>> print(f"Solution: {model.g_best.solution}, Fitness: {model.g_best.target.fitness}")

References

[1] Awad, N. H., Ali, M. Z., & Suganthan, P. N. (2017, June). Ensemble sinusoidal differential covariance matrix adaptation with Euclidean neighborhood for solving CEC2017 benchmark problems. In 2017 IEEE congress on evolutionary computation (CEC) (pp. 372-379). IEEE.

before_main_loop()

binomial_crossover(target, mutant, CR=None)

Standard binomial crossover

covariance_matrix_crossover(target, mutant)

Covariance matrix learning with Euclidean neighborhood

current_to_pbest_mutation(idx, F, p=0.1)

Current-to-pbest/1 mutation strategy

evolve(epoch)

The main operations (equations) of algorithm. Inherit from Optimizer class

Parameters

epoch (int) – The current iteration

initialize_variables()

linear_population_reduction(epoch, max_epoch)

Linear population size reduction

sinusoidal_adaptation(epoch, max_epoch, config_type, freq=None)

Sinusoidal parameter adaptation config_type: 1 for non-adaptive decreasing, 2 for adaptive increasing

update_sinusoidal_probabilities(epoch)

Update probabilities for sinusoidal configurations

weighted_lehmer_mean(S_values, delta_f)

Calculate weighted Lehmer mean