hello_nsga3_pymoo.py

#%%
import random
import numpy as np
from pymoo.core.problem            import ElementwiseProblem
from pymoo.algorithms.moo.nsga2    import NSGA2
from pymoo.algorithms.moo.nsga3    import NSGA3
from pymoo.optimize                import minimize
from pymoo.util.ref_dirs           import get_reference_directions

from pymoo.operators.sampling.rnd  import FloatRandomSampling
from pymoo.operators.crossover.sbx import SBX
from pymoo.operators.mutation.pm   import PM
from pymoo.termination             import get_termination


#%%


#%%
MAFAST_LOWER  = 10.0
MAFAST_UPPER  = 200.0

MASLOW_LOWER  = 50.0
MASLOW_UPPER  = 450.0

PCT_EPS_LOWER = 0.05
PCT_EPS_UPPER = 10.0


class HelloWorld(ElementwiseProblem):
    def __init__(self, **kwargs):
        super().__init__(
            n_var     = 3, 
            n_obj     = 2, 
            xl        = np.array([MAFAST_LOWER, MASLOW_LOWER, PCT_EPS_LOWER]), 
            xu        = np.array([MAFAST_UPPER, MASLOW_UPPER, PCT_EPS_UPPER]), 
            **kwargs)

    def _evaluate(self, x, out, *args, **kwargs):
        print(f"evaluation : {x[0]} {x[1]} {x[2]}")
        fitnesses = [
            random.choice(list(np.random.uniform(size=100, low=0.01, high=10.01))),
            random.choice(list(np.random.uniform(size=300, low=0.05, high=20.01))),
        ]
        print(f"fitnesses {fitnesses}")
        out["F"] = fitnesses

problem = HelloWorld()


#%%
ref_dirs = get_reference_directions("das-dennis", 2, n_partitions=12)
algorithm = NSGA3(
    pop_size     = 100,
    ref_dirs     = ref_dirs,
    sampling     = FloatRandomSampling(),
    crossover    = SBX(prob=0.9, eta=15),
    mutation     = PM(eta=20),
    termination  = get_termination("n_gen", 40),
    eliminate_duplicates=True
    )


#%%
res = minimize(
    problem,
    algorithm,
    ('n_gen', 10),
    seed=1,
    verbose=True
)

#print('hash', res)


#%%


#%%


#%%