"Legacy code. To be updated or depreciated."
from pyrvea.EAs.RVEA import RVEA
from pyrvea.OtherTools.ReferenceVectors import ReferenceVectors
from typing import TYPE_CHECKING
if TYPE_CHECKING:
from pyrvea.Population.Population import Population
[docs]class slowRVEA(RVEA):
"""RVEA variant that impliments slow reference vector movement."""
def __init__(self, population: "Population", ea_parameters):
"""Initialize a Base Decomposition EA.
This will call methods to set up the parameters of RVEA, create
Reference Vectors, and (as of Feb 2019) run the first iteration of RVEA.
Parameters
----------
population : "Population"
This variable is updated as evolution takes place
EA_parameters : dict
Takes the EA parameters
Returns
-------
Population:
Returns the Population after evolution.
"""
if ea_parameters:
self.params = self.set_params(population, **ea_parameters)
else:
self.params = self.set_params(population)
# if population.individuals.shape[0] == 0:
# create_new_individuals(pop_size=self.params["population_size"])
# # print("Using BaseDecompositionEA init")
# self._next_iteration(population)
[docs] def set_params(
self,
population: "Population",
generations_per_iteration: int = 10,
iterations: int = 10,
Alpha: float = 2,
ref_point: list = None,
old_point: list = None,
**kwargs
):
"""Set up the parameters. Save in RVEA.params. Note, this should be
changed to align with the current structure.
Parameters
----------
population : Population
Population object
Alpha : float
The alpha parameter of APD selection.
plotting : bool
Useless really.
Returns
-------
"""
ref_vectors = ReferenceVectors(
number_of_objectives=population.problem.num_of_objectives,
creation_type="Sparse_Focused",
ref_point=old_point,
)
if ref_point is None:
ref_point = ref_vectors.values[0]
rveaparams = {
"reference_vectors": ref_vectors,
"population_size": ref_vectors.number_of_vectors,
"generations": generations_per_iteration,
"iterations": iterations,
"Alpha": Alpha,
"current_iteration_gen_count": 0,
"current_iteration_count": 0,
"current_total_gen_count": 0,
"total_generations": iterations * generations_per_iteration,
"ref_point": ref_point,
}
rveaparams.update(kwargs)
return rveaparams
[docs] def _run_interruption(self, population: "Population"):
self.params["reference_vectors"].slow_interactive_adapt(
self.params["ref_point"]
)
