def test_rosenbrock(self):
bounds = np.zeros((2, 2))
bounds[:, 0] = -3.0
bounds[:, 1] = 3.0
np.random.seed(1234124)
my_opt = sbopt.RbfOpt(my_fun, bounds, polish=True)
x, y, _, _ = my_opt.minimize(verbose=0)
self.assertTrue(np.isclose(y, 0.0, rtol=1e-4, atol=1e-4))
self.assertTrue(np.isclose(x[0], 1.0, rtol=1e-2, atol=1e-2))
self.assertTrue(np.isclose(x[1], 1.0, rtol=1e-2, atol=1e-2))
def test_rosenbrock_distance_ei(self):
bounds = np.zeros((2, 2))
bounds[:, 0] = -3.0
bounds[:, 1] = 3.0
np.random.seed(1234124)
my_opt = sbopt.RbfOpt(my_fun,
bounds,
polish=False,
n_local_optimze=2,
acquisition='distance',
exploration='distance',
rbf_function='multiquadric')
x, y, _, _ = my_opt.minimize(verbose=0)
self.assertTrue(np.isclose(y, 0.0, rtol=1e-4, atol=1e-4))
self.assertTrue(np.isclose(x[0], 1.0, rtol=1e-2, atol=1e-2))
self.assertTrue(np.isclose(x[1], 1.0, rtol=1e-2, atol=1e-2))
bounds = np.zeros((2, 2))
bounds[:, 0] = -3.0
bounds[:, 1] = 3.0
# set random seed for reproducibility
np.random.seed(1234124)
# initialize the RbfOpt object
my_opt = sbopt.RbfOpt(
my_fun, # your objective function to minimize
bounds, # bounds for your design variables
initial_design='latin', # initial design type
# 'latin' default, or 'random'
initial_design_ndata=20, # number of initial points
n_local_optimze=20, # number of local BFGS optimizers
# scipy radial basis function parameters see
# https://docs.scipy.org/doc/scipy/reference/generated/scipy.interpolate.Rbf.html
rbf_function='linear', # default for SbOpt is 'linear'
# while the default for scipy.interpolate.Rbf is
# 'multi-quadratic'
epsilon=None, # (default)
smooth=0.0, # (default)
norm='euclidean' # (default)
)
# run the optimizer
result = my_opt.minimize(
max_iter=100, # maximum number of iterations
# (default)
n_same_best=20, # number of iterations to run
# without improving best function value (default)
eps=1e-6, # minimum distance a new design point
# may be from an existing design point (default)
r = 6.0
s = 10.0
t = 1.0 / (8.0*np.pi)
def my_fun(x):
# define the Branin-Hoo function to minmize
A = a*(x[1] - (b*x[0]**2) + c*x[0] - r)**2
B = s*(1.0 - t)*np.cos(x[0])
return A + B + s
bounds = np.zeros((2, 2))
bounds[0, 0] = -5.0
bounds[0, 1] = 10.0
bounds[1, 1] = 15.0
# set random seed for reproducibility
np.random.seed(1234124)
# initialize the RbfOpt object
my_opt = sbopt.RbfOpt(my_fun, # your objective function to minimize
bounds, # bounds for your design variables
)
# run the optimizer
result = my_opt.minimize()
print('Best design variables:', result[0])
print('Best function value:', result[1])
print('Convergence by max iteration:', result[2])
print('Convergence by n_same_best:', result[3])
t = 1.0 / (8.0*np.pi)
def my_fun(x):
# define the Branin-Hoo function to minmize
A = a*(x[1] - (b*x[0]**2) + c*x[0] - r)**2
B = s*(1.0 - t)*np.cos(x[0])
return A + B + s
bounds = np.zeros((2, 2))
bounds[0, 0] = -5.0
bounds[0, 1] = 10.0
bounds[1, 1] = 15.0
# set random seed for reproducibility
np.random.seed(1234124)
# initialize the RbfOpt object
my_opt = sbopt.RbfOpt(my_fun, # your objective function to minimize
bounds, # bounds for your design variables
n_local_optimze=5,
initial_design_ndata=5)
# run the optimizer
result = my_opt.minimize(strategy='all_local_reflect', eps=1e-3,
max_iter=100)
print('Best design variables:', result[0])
print('Best function value:', result[1])
print('Convergence by max iteration:', result[2])
print('Convergence by n_same_best:', result[3])