f_rosenbrock.py

import numpy as np
from experiments.objective_functions.f_base import BaseObjective


class Rosenbrock(BaseObjective):
    def __init__(self, d, int_opt=(-3., 3.)):
        super(Rosenbrock, self).__init__(d, int_opt=int_opt)
        self.c = np.maximum(1, np.sqrt(self.d) / 8)

    def evaluate_full(self, x):
        x = np.atleast_2d(x)
        assert x.shape[1] == self.d

        z = self.c * (x - self.x_opt) + 1
        a = z[:, :-1]**2 - z[:, 1:]
        b = z[:, :-1] - 1

        out = np.sum(100 * a**2 + b**2, axis=1) + self.f_opt

        return out


class RosenbrockRotated(BaseObjective):
    def __init__(self, d, int_opt=(-3., 3.)):
        super(RosenbrockRotated, self).__init__(d, int_opt=int_opt)
        self.c = np.maximum(1, np.sqrt(self.d) / 8)

    def evaluate_full(self, x):
        x = np.atleast_2d(x)
        assert x.shape[1] == self.d

        z = (self.c * self.r @ x.T + 1/2).T
        a = z[:, :-1]**2 - z[:, 1:]
        b = z[:, :-1] - 1

        out = np.sum(100 * a**2 + b**2, axis=1) + self.f_opt

        return out


class RosenbrockRaw(BaseObjective):
    def __init__(self, d, int_opt=(-3., 3.)):
        super(RosenbrockRaw, self).__init__(d, int_opt=int_opt)
        self.x_opt = np.ones((1, d))
        self.f_opt = 0

    def evaluate_full(self, x):
        x = np.atleast_2d(x)
        assert x.shape[1] == self.d

        a = x[:, :-1]**2 - x[:, 1:]
        b = x[:, :-1] - 1

        out = np.sum(100 * a**2 + b**2, axis=1)

        return out