def generate_input_data(n_simulations, method = "random"):
"Generate random points x1 and x2 for evaluating the multivalued 2D Branin function"
n_simulations = int(n_simulations)
assert(n_simulations > 0)
assert method == "random" or method == "lhd"
if method == "random":
ed = MonteCarloDesign([uniform(loc = -5., scale = 15.).ppf, uniform(loc = 0., scale = 15.).ppf])
elif method == "lhd":
ed = LatinHypercubeDesign([uniform(loc = -5., scale = 15.).ppf, uniform(loc = 0., scale = 15.).ppf])
inputs = ed.sample(n_simulations)
return inputs
def run_model(n_simulations, n_testing):
"Generate training data, fit emulator, and test model accuracy on random points, returning RMSE"
# run MICE Model
print('running MICE')
ed = LatinHypercubeDesign(design_space)
n_init = 5
md = MICEDesign(ed,
f,
n_samples=n_simulations - n_init,
n_init=n_init,
n_cand=100)
md.run_sequential_design()
inputs_mice = md.get_inputs()
targets_mice = md.get_targets()
print('fitting GPs')
gp_mice = GaussianProcess(inputs_mice, np.squeeze(targets_mice))
gp_mice = fit_GP_MAP(gp_mice)
# run LHD model
inputs, targets = generate_training_data(n_simulations)
gp = GaussianProcess(inputs, targets)
gp = fit_GP_MAP(gp)
print("making predictions")
testing, test_targets = generate_test_data(n_testing)
norm_const = np.max(test_targets) - np.min(test_targets)
test_vals_mice, unc_mice, deriv = gp_mice.predict(testing,
deriv=False,
unc=True)
test_vals, unc, deriv = gp.predict(testing, deriv=False, unc=True)
return (
np.sqrt(np.sum(
(test_vals - test_targets)**2) / float(n_testing)) / norm_const,
np.sqrt(np.sum(unc**2) / float(n_testing)) / norm_const**2,
np.sqrt(np.sum(
(test_vals_mice - test_targets)**2) / float(n_testing)) /
norm_const,
np.sqrt(np.sum(unc_mice**2) / float(n_testing)) / norm_const**2)
def generate_input_data(n_simulations, n_dimensions, method="random"):
"Generate n_simulations evaluation points the n_dimensions dimensional Rosenbrock function"
n_simulations = int(n_simulations)
n_dimensions = int(n_dimensions)
assert (n_simulations > 0)
assert (n_dimensions > 0)
assert method == "random" or method == "lhd"
if method == "random":
ed = MonteCarloDesign(n_dimensions, (-5., 10.))
elif method == "lhd":
ed = LatinHypercubeDesign(n_dimensions, (-5., 10.))
inputs = ed.sample(n_simulations)
return inputs