EDAspy.optimization.custom.initialization_models package

Submodules

EDAspy.optimization.custom.initialization_models.categorical_geninit module

class EDAspy.optimization.custom.initialization_models.categorical_geninit.CategoricalSampling(n_variables: int, possible_values: List | array, frequency: List | array)[source]

Bases: GenInit

Initial generation simulator based on the Latin Hypercube Sampling process.

sample(size: int) → array[source]

Sample several times the initializer.

Parameters:

size – number of samplings.

Returns:

array with the dataset sampled

Return type:

np.array

EDAspy.optimization.custom.initialization_models.latin_hypercube_sampling_geninit module

class EDAspy.optimization.custom.initialization_models.latin_hypercube_sampling_geninit.LatinHypercubeSampling(n_variables: int, lower_bound: array | List[float] | float = -100, upper_bound: List[float] | float = 100)[source]

Bases: GenInit

Initial generation simulator based on the Latin Hypercube Sampling process.

sample(size: int, post_process: bool = False) → array[source]

Sample several times the initializer.

Parameters:

  • size – number of samplings.

  • post_process – Post processing to ensure diversity between solutions using Lloyd-Max algorithm.

Returns:

array with the dataset sampled

Return type:

np.array

EDAspy.optimization.custom.initialization_models.multi_gauss_geninit module

class EDAspy.optimization.custom.initialization_models.multi_gauss_geninit.MultiGaussGenInit(n_variables: int, means_vector: array = array([], dtype=float64), cov_matrix: array = array([], dtype=float64), lower_bound: float = -100, upper_bound: float = 100)[source]

Bases: GenInit

Initial generation simulator based on the probabilistic model of multivariate Gaussian distribution.

sample(size: int) → array[source]

Sample several times the initializator.

Parameters:

size – number of samplings.

Returns:

array with the dataset sampled.

Return type:

np.array

EDAspy.optimization.custom.initialization_models.uni_bin_geninit module

class EDAspy.optimization.custom.initialization_models.uni_bin_geninit.UniBinGenInit(n_variables: int, means_vector: array = array([], dtype=float64))[source]

Bases: GenInit

Initial generation simulator based on the probabilistic model of univariate binary probabilities.

sample(size: int) → array[source]

Sample several times the initializator.

Parameters:

size – number of samplings.

Returns:

array with the dataset sampled

Return type:

np.array

EDAspy.optimization.custom.initialization_models.uni_gauss_geninit module

class EDAspy.optimization.custom.initialization_models.uni_gauss_geninit.UniGaussGenInit(n_variables: int, means_vector: array = array([], dtype=float64), stds_vector: array = array([], dtype=float64), lower_bound: int = -100, higher_bound: int = 100)[source]

Bases: GenInit

Initial generation simulator based on the probabilistic model of univariate binary probabilities.

sample(size) → array[source]

Sample several times the initializator.

Parameters:

size – number of samplings.

Returns:

array with the dataset sampled.

Return type:

np.array.

EDAspy.optimization.custom.initialization_models.uniform_geninit module

class EDAspy.optimization.custom.initialization_models.uniform_geninit.UniformGenInit(n_variables: int, lower_bound: array | List[float] | float = -100, upper_bound: List[float] | float = 100)[source]

Bases: GenInit

Initial generation simulator based on independent uniform distributions.

sample(size: int) → array[source]

Sample several times the initializator.

Parameters:

size – number of samplings.

Returns:

array with the dataset sampled.

Return type:

np.array.

Module contents