def test_sin_metamodel(self):
# create a MetaModel for sine and add it to a Problem
sin_mm = MetaModel()
sin_mm.add_input('x', 0.)
sin_mm.add_output('f_x', 0.)
prob = Problem()
prob.model.add_subsystem('sin_mm', sin_mm)
# check that missing surrogate is detected in check_config
testlogger = TestLogger()
prob.setup(logger=testlogger)
# Conclude setup but don't run model.
prob.final_setup()
msg = ("No default surrogate model is defined and the "
"following outputs do not have a surrogate model:\n"
"['f_x']\n"
"Either specify a default_surrogate, or specify a "
"surrogate model for all outputs.")
self.assertEqual(len(testlogger.get('error')), 1)
self.assertTrue(msg in testlogger.get('error')[0])
# check that output with no specified surrogate gets the default
sin_mm.default_surrogate = FloatKrigingSurrogate()
prob.setup(check=False)
surrogate = sin_mm._metadata('f_x').get('surrogate')
self.assertTrue(isinstance(surrogate, FloatKrigingSurrogate),
'sin_mm.f_x should get the default surrogate')
# check error message when no training data is provided
with self.assertRaises(RuntimeError) as cm:
prob.run_model()
msg = ("MetaModel: The following training data sets must be "
"provided as metadata for sin_mm: ['train:x', 'train:f_x']")
self.assertEqual(str(cm.exception), msg)
# train the surrogate and check predicted value
sin_mm.metadata['train:x'] = np.linspace(0, 10, 20)
sin_mm.metadata['train:f_x'] = .5 * np.sin(sin_mm.metadata['train:x'])
prob['sin_mm.x'] = 2.1
prob.run_model()
assert_rel_error(self, prob['sin_mm.f_x'],
.5 * np.sin(prob['sin_mm.x']), 1e-4)
def test_sin_metamodel_preset_data(self):
# preset training data
x = np.linspace(0, 10, 200)
f_x = .5 * np.sin(x)
# create a MetaModel for Sin and add it to a Problem
sin_mm = MetaModel()
sin_mm.add_input('x', 0., training_data=np.linspace(0, 10, 200))
sin_mm.add_output('f_x', 0., training_data=f_x)
prob = Problem()
prob.model.add_subsystem('sin_mm', sin_mm)
# check that missing surrogate is detected in check_setup
testlogger = TestLogger()
prob.setup(logger=testlogger)
# Conclude setup but don't run model.
prob.final_setup()
msg = ("No default surrogate model is defined and the "
"following outputs do not have a surrogate model:\n"
"['f_x']\n"
"Either specify a default_surrogate, or specify a "
"surrogate model for all outputs.")
self.assertEqual(len(testlogger.get('error')), 1)
self.assertTrue(msg in testlogger.get('error')[0])
# check that output with no specified surrogate gets the default
sin_mm.default_surrogate = FloatKrigingSurrogate()
prob.setup(check=False)
surrogate = sin_mm._metadata('f_x').get('surrogate')
self.assertTrue(isinstance(surrogate, FloatKrigingSurrogate),
'sin_mm.f_x should get the default surrogate')
prob['sin_mm.x'] = 2.22
prob.run_model()
assert_rel_error(self, prob['sin_mm.f_x'],
.5 * np.sin(prob['sin_mm.x']), 1e-4)
def test_sin_metamodel_rmse(self):
# create MetaModel with Kriging, using the rmse option
sin_mm = MetaModel()
sin_mm.add_input('x', 0.)
sin_mm.add_output('f_x', 0.)
sin_mm.default_surrogate = KrigingSurrogate(eval_rmse=True)
# add it to a Problem
prob = Problem()
prob.model.add_subsystem('sin_mm', sin_mm)
prob.setup(check=False)
# train the surrogate and check predicted value
sin_mm.metadata['train:x'] = np.linspace(0,10,20)
sin_mm.metadata['train:f_x'] = np.sin(sin_mm.metadata['train:x'])
prob['sin_mm.x'] = 2.1
prob.run_model()
assert_rel_error(self, prob['sin_mm.f_x'], np.sin(2.1), 1e-4) # mean
self.assertTrue(self, sin_mm._metadata('f_x')['rmse'] < 1e-5) # std deviation
def test_basics(self):
# create a metamodel component
mm = MetaModel()
mm.add_input('x1', 0.)
mm.add_input('x2', 0.)
mm.add_output('y1', 0.)
mm.add_output('y2', 0., surrogate=FloatKrigingSurrogate())
mm.default_surrogate = ResponseSurface()
# add metamodel to a problem
prob = Problem(model=Group())
prob.model.add_subsystem('mm', mm)
prob.setup(check=False)
# check that surrogates were properly assigned
surrogate = mm._metadata('y1').get('surrogate')
self.assertTrue(isinstance(surrogate, ResponseSurface))
surrogate = mm._metadata('y2').get('surrogate')
self.assertTrue(isinstance(surrogate, FloatKrigingSurrogate))
# populate training data
mm.metadata['train:x1'] = [1.0, 2.0, 3.0]
mm.metadata['train:x2'] = [1.0, 3.0, 4.0]
mm.metadata['train:y1'] = [3.0, 2.0, 1.0]
mm.metadata['train:y2'] = [1.0, 4.0, 7.0]
# run problem for provided data point and check prediction
prob['mm.x1'] = 2.0
prob['mm.x2'] = 3.0
self.assertTrue(mm.train) # training will occur before 1st run
prob.run_model()
assert_rel_error(self, prob['mm.y1'], 2.0, .00001)
assert_rel_error(self, prob['mm.y2'], 4.0, .00001)
# run problem for interpolated data point and check prediction
prob['mm.x1'] = 2.5
prob['mm.x2'] = 3.5
self.assertFalse(mm.train) # training will not occur before 2nd run
prob.run_model()
assert_rel_error(self, prob['mm.y1'], 1.5934, .001)
# change default surrogate, re-setup and check that metamodel re-trains
mm.default_surrogate = FloatKrigingSurrogate()
prob.setup(check=False)
surrogate = mm._metadata('y1').get('surrogate')
self.assertTrue(isinstance(surrogate, FloatKrigingSurrogate))
self.assertTrue(mm.train) # training will occur after re-setup
mm.warm_restart = True # use existing training data
prob['mm.x1'] = 2.5
prob['mm.x2'] = 3.5
prob.run_model()
assert_rel_error(self, prob['mm.y1'], 1.5, 1e-2)
