def test_array_multi_vectorize(self):
def branin(x):
x1 = 15 * x[0] - 5
x2 = 15 * x[1]
return (x2 - (5.1 / (4. * np.pi**2.)) * x1**2. + 5. * x1 / np.pi -
6.)**2. + 10. * (1. - 1. / (8. * np.pi)) * np.cos(x1) + 10.
# Add a linear error
def branin_low_fidelity(x):
return branin(x) + 30. * x[1] + 10.
mm = om.MultiFiMetaModelUnStructuredComp(nfi=2)
mm.add_input('x', np.zeros((1, 2)))
mm.add_output('y', np.zeros((1, )))
mm.options['default_surrogate'] = om.MultiFiCoKrigingSurrogate(
normalize=False)
prob = om.Problem()
prob.model.add_subsystem('mm', mm)
prob.setup()
x = [
[
[0.13073587, 0.24909577], # expensive (hifi) doe
[0.91915571, 0.4735261],
[0.75830543, 0.13321705],
[0.51760477, 0.34594101],
[0.03531219, 0.77765831],
[0.27249206, 0.5306115],
[0.62762489, 0.65778471],
[0.3914706, 0.09852519],
[0.86565585, 0.85350002],
[0.40806563, 0.91465314]
],
[
[0.91430235, 0.17029894], # cheap (lowfi) doe
[0.99329651, 0.76431519],
[0.2012252, 0.35006032],
[0.61707854, 0.90210676],
[0.15113004, 0.0133355],
[0.07108082, 0.55344447],
[0.4483159, 0.52182902],
[0.5926638, 0.06595122],
[0.66305449, 0.48579608],
[0.47965045, 0.7407793],
[0.13073587, 0.24909577], # notice hifi doe inclusion
[0.91915571, 0.4735261],
[0.75830543, 0.13321705],
[0.51760477, 0.34594101],
[0.03531219, 0.77765831],
[0.27249206, 0.5306115],
[0.62762489, 0.65778471],
[0.3914706, 0.09852519],
[0.86565585, 0.85350002],
[0.40806563, 0.91465314]
]
]
y = np.array([[branin(case) for case in x[0]],
[branin_low_fidelity(case) for case in x[1]]])
def test_surrogate_message_format(self):
mm = om.MultiFiMetaModelUnStructuredComp(nfi=2)
mm.add_input('x', np.zeros((1, 2)))
mm.add_output('y', np.zeros((1, )))
mm.options['default_surrogate'] = om.MultiFiCoKrigingSurrogate(normalize=False)
prob = om.Problem()
prob.model.add_subsystem('mm', mm)
prob.setup()
x = [[[ 0.13073587, 0.24909577], # expensive (hifi) doe
[ 0.13073587, 0.24909577],
[ 0.40806563, 0.91465314]],
[[ 0.91430235, 0.17029894], # cheap (lowfi) doe
[ 0.40806563, 0.91465314]]]
mm.options['train:x'] = x[0]
mm.options['train:y'] = np.array([1, 2, 3])
mm.options['train:x_fi2'] = x[1]
mm.options['train:y_fi2'] = np.array([1, 2])
prob['mm.x'] = np.array([[2./3., 1./3.]])
with self.assertRaises(ValueError) as cm:
prob.run_model()
expected = ("mm: Multiple input features cannot have the same value.")
self.assertEqual(str(cm.exception), expected)
def test_om_slice_in_add_input(self):
mm = om.MultiFiMetaModelUnStructuredComp(nfi=2)
mm.add_input('x', np.ones(3), src_indices=om.slicer[:, 1])
mm.add_output('y', np.zeros((1, )))
mm.options['default_surrogate'] = om.MultiFiCoKrigingSurrogate(
normalize=False)
arr = np.array([[1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4]])
prob = om.Problem()
prob.model.add_subsystem('mm', mm)
prob.model.add_subsystem('indep', om.IndepVarComp('x', arr))
prob.model.connect('indep.x', 'mm.x')
prob.setup()
assert_near_equal(prob['mm.x'],
np.array([[1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4]]))
def test_2_input_2_fidelity(self):
import numpy as np
import openmdao.api as om
mm = om.MultiFiMetaModelUnStructuredComp(nfi=2)
mm.add_input('x', np.zeros((1, 2)))
mm.add_output('y', np.zeros((1, )))
# Surrrogate model that implements the multifidelity cokriging method.
mm.options['default_surrogate'] = om.MultiFiCoKrigingSurrogate(
normalize=False)
prob = om.Problem()
prob.model.add_subsystem('mm', mm)
prob.setup()
x_hifi = np.array([
[0.13073587, 0.24909577], # expensive (hifi) doe
[0.91915571, 0.4735261],
[0.75830543, 0.13321705],
[0.51760477, 0.34594101],
[0.03531219, 0.77765831],
[0.27249206, 0.5306115],
[0.62762489, 0.65778471],
[0.3914706, 0.09852519],
[0.86565585, 0.85350002],
[0.40806563, 0.91465314]
])
y_hifi = np.array([
69.22687251161716, 28.427292491743817, 20.36175030334259,
7.840766670948326, 23.950783505007422, 16.0326610719367,
77.32264403894713, 26.625242780670835, 135.85615334210993,
101.43980212355875
])
x_lofi = np.array([
[0.91430235, 0.17029894], # cheap (lowfi) doe
[0.99329651, 0.76431519],
[0.2012252, 0.35006032],
[0.61707854, 0.90210676],
[0.15113004, 0.0133355],
[0.07108082, 0.55344447],
[0.4483159, 0.52182902],
[0.5926638, 0.06595122],
[0.66305449, 0.48579608],
[0.47965045, 0.7407793],
[0.13073587, 0.24909577], # notice hifi doe inclusion
[0.91915571, 0.4735261],
[0.75830543, 0.13321705],
[0.51760477, 0.34594101],
[0.03531219, 0.77765831],
[0.27249206, 0.5306115],
[0.62762489, 0.65778471],
[0.3914706, 0.09852519],
[0.86565585, 0.85350002],
[0.40806563, 0.91465314]
])
y_lofi = list([
18.204898470295255, 107.66640600958577, 46.11717344625053,
186.002239934648, 135.12480249921992, 65.3605467926758,
51.72316385370553, 15.541873662737451, 72.77648156410065,
100.33324800434931, 86.69974561161716, 52.63307549174382,
34.358261803342586, 28.218996970948325, 57.280532805007425,
41.9510060719367, 107.05618533894713, 39.580998480670836,
171.46115394210995, 138.87939632355875
])
mm.options['train:x'] = x_hifi
mm.options['train:y'] = y_hifi
mm.options['train:x_fi2'] = x_lofi
mm.options['train:y_fi2'] = y_lofi
prob.set_val('mm.x', np.array([[2. / 3., 1. / 3.]]))
prob.run_model()
assert_near_equal(prob.get_val('mm.y'), 26.26, tolerance=0.02)
prob['mm.x'] = np.array([[1. / 3., 2. / 3.]])
prob.run_model()
assert_near_equal(prob['mm.y'], 36.1031735, tolerance=0.02)
# Now, vectorized model with both points predicted together.
mm = om.MultiFiMetaModelUnStructuredComp(nfi=2, vec_size=2)
mm.add_input('x', np.zeros((2, 1, 2)))
mm.add_output('y', np.zeros((
2,
1,
)))
mm.options['default_surrogate'] = om.MultiFiCoKrigingSurrogate(
normalize=False)
prob = om.Problem()
prob.model.add_subsystem('mm', mm)
prob.setup()
mm.options['train:x'] = x[0]
mm.options['train:y'] = y[0]
mm.options['train:x_fi2'] = x[1]
mm.options['train:y_fi2'] = y[1]
prob['mm.x'] = np.array([[[2. / 3., 1. / 3.]], [[1. / 3., 2. / 3.]]])
prob.run_model()
assert_near_equal(prob['mm.y'], [[26.26], [36.1031735]],
tolerance=0.02)
