def test_KnowledgeGradient(self):
model = KnowledgeGradient()
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
feature_names=self.feature_names,
metric_names=self.metric_names,
task_features=[],
fidelity_features=[],
)
n = 2
X_dummy = torch.rand(1, n, 4, dtype=self.dtype, device=self.device)
acq_dummy = torch.tensor(0.0, dtype=self.dtype, device=self.device)
with mock.patch(self.optimize_acqf) as mock_optimize_acqf:
mock_optimize_acqf.side_effect = [(X_dummy, acq_dummy)]
Xgen, wgen, _, __ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": self.optimizer_options,
},
)
self.assertTrue(torch.equal(Xgen, X_dummy.cpu()))
self.assertTrue(torch.equal(wgen, torch.ones(n, dtype=self.dtype)))
# called once, the best point call is not caught by mock
mock_optimize_acqf.assert_called_once()
ini_dummy = torch.rand(10, 32, 3, dtype=self.dtype, device=self.device)
optimizer_options2 = {
"num_restarts": 1,
"raw_samples": 1,
"maxiter": 5,
"batch_limit": 1,
"partial_restarts": 2,
}
with mock.patch(
"ax.models.torch.botorch_kg.gen_one_shot_kg_initial_conditions",
return_value=ini_dummy,
) as mock_warmstart_initialization:
Xgen, wgen, _, __ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": optimizer_options2,
},
)
mock_warmstart_initialization.assert_called_once()
obj = ScalarizedObjective(weights=self.objective_weights)
dummy_acq = PosteriorMean(model=model.model, objective=obj)
with mock.patch(
"ax.models.torch.utils.PosteriorMean", return_value=dummy_acq
) as mock_posterior_mean:
Xgen, wgen, _, __ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": optimizer_options2,
},
)
self.assertEqual(mock_posterior_mean.call_count, 2)
# Check best point selection within bounds (some numerical tolerance)
xbest = model.best_point(
bounds=self.bounds, objective_weights=self.objective_weights
)
lb = torch.tensor([b[0] for b in self.bounds]) - 1e-5
ub = torch.tensor([b[1] for b in self.bounds]) + 1e-5
self.assertTrue(torch.all(xbest <= ub))
self.assertTrue(torch.all(xbest >= lb))
# test error message
linear_constraints = (
torch.tensor([[0.0, 0.0, 0.0], [0.0, 1.0, 0.0]]),
torch.tensor([[0.5], [1.0]]),
)
with self.assertRaises(UnsupportedError):
Xgen, wgen = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=linear_constraints,
)
# test input warping
self.assertFalse(model.use_input_warping)
model = KnowledgeGradient(use_input_warping=True)
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
feature_names=self.feature_names,
metric_names=self.metric_names,
task_features=[],
fidelity_features=[],
)
self.assertTrue(model.use_input_warping)
self.assertTrue(hasattr(model.model, "input_transform"))
self.assertIsInstance(model.model.input_transform, Warp)
# test loocv pseudo likelihood
self.assertFalse(model.use_loocv_pseudo_likelihood)
model = KnowledgeGradient(use_loocv_pseudo_likelihood=True)
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
feature_names=self.feature_names,
metric_names=self.metric_names,
task_features=[],
fidelity_features=[],
)
self.assertTrue(model.use_loocv_pseudo_likelihood)
def test_KnowledgeGradient(self):
model = KnowledgeGradient()
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
feature_names=self.feature_names,
metric_names=self.metric_names,
task_features=[],
fidelity_features=[],
)
n = 2
best_point_dummy = torch.rand(1,
3,
dtype=self.dtype,
device=self.device)
X_dummy = torch.rand(1, n, 4, dtype=self.dtype, device=self.device)
acq_dummy = torch.tensor(0.0, dtype=self.dtype, device=self.device)
with mock.patch(self.optimize_acqf) as mock_optimize_acqf:
mock_optimize_acqf.side_effect = [
(best_point_dummy, None),
(X_dummy, acq_dummy),
]
Xgen, wgen, _ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": self.optimizer_options,
},
)
self.assertTrue(torch.equal(Xgen, X_dummy.cpu()))
self.assertTrue(torch.equal(wgen, torch.ones(n, dtype=self.dtype)))
mock_optimize_acqf.assert_called(
) # called twice, once for best_point
ini_dummy = torch.rand(10, 32, 3, dtype=self.dtype, device=self.device)
optimizer_options2 = {
"num_restarts": 1,
"raw_samples": 1,
"maxiter": 5,
"batch_limit": 1,
"partial_restarts": 2,
}
with mock.patch(
"ax.models.torch.botorch_kg.gen_one_shot_kg_initial_conditions",
return_value=ini_dummy,
) as mock_warmstart_initialization:
Xgen, wgen, _ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": optimizer_options2,
},
)
mock_warmstart_initialization.assert_called_once()
obj = ScalarizedObjective(weights=self.objective_weights)
dummy_acq = PosteriorMean(model=model.model, objective=obj)
with mock.patch("ax.models.torch.botorch_kg.PosteriorMean",
return_value=dummy_acq) as mock_posterior_mean:
Xgen, wgen, _ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": optimizer_options2,
},
)
self.assertEqual(mock_posterior_mean.call_count, 2)
# Check best point selection
X_dummy = torch.rand(3)
acq_dummy = torch.tensor(0.0)
with mock.patch(self.optimize_acqf,
return_value=(X_dummy,
acq_dummy)) as mock_optimize_acqf:
xbest = model.best_point(bounds=self.bounds,
objective_weights=self.objective_weights)
self.assertTrue(torch.equal(xbest, X_dummy))
mock_optimize_acqf.assert_called_once()
# test error message
linear_constraints = (
torch.tensor([[0.0, 0.0, 0.0], [0.0, 1.0, 0.0]]),
torch.tensor([[0.5], [1.0]]),
)
with self.assertRaises(UnsupportedError):
Xgen, wgen = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=linear_constraints,
)
def test_KnowledgeGradient_multifidelity(self):
model = KnowledgeGradient()
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
task_features=[],
feature_names=self.feature_names,
metric_names=["L2NormMetric"],
fidelity_features=[2],
)
# Check best point selection within bounds (some numerical tolerance)
xbest = model.best_point(
bounds=self.bounds,
objective_weights=self.objective_weights,
target_fidelities={2: 5.0},
)
lb = torch.tensor([b[0] for b in self.bounds]) - 1e-5
ub = torch.tensor([b[1] for b in self.bounds]) + 1e-5
self.assertTrue(torch.all(xbest <= ub))
self.assertTrue(torch.all(xbest >= lb))
# check error when no target fidelities are specified
with self.assertRaises(RuntimeError):
model.best_point(
bounds=self.bounds, objective_weights=self.objective_weights
)
# check generation
n = 2
X_dummy = torch.zeros(1, n, 3, dtype=self.dtype, device=self.device)
acq_dummy = torch.tensor(0.0, dtype=self.dtype, device=self.device)
dummy = (X_dummy, acq_dummy)
with mock.patch(self.optimize_acqf, side_effect=[dummy]) as mock_optimize_acqf:
Xgen, wgen, _, __ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": self.optimizer_options,
},
target_fidelities={2: 5.0},
)
self.assertTrue(torch.equal(Xgen, X_dummy.cpu()))
self.assertTrue(torch.equal(wgen, torch.ones(n, dtype=self.dtype)))
mock_optimize_acqf.assert_called() # called twice, once for best_point
# test error message
linear_constraints = (
torch.tensor([[0.0, 0.0, 0.0], [0.0, 1.0, 0.0]]),
torch.tensor([[0.5], [1.0]]),
)
with self.assertRaises(UnsupportedError):
xbest = model.best_point(
bounds=self.bounds,
linear_constraints=linear_constraints,
objective_weights=self.objective_weights,
target_fidelities={2: 1.0},
)
# test input warping
self.assertFalse(model.use_input_warping)
model = KnowledgeGradient(use_input_warping=True)
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
task_features=[],
feature_names=self.feature_names,
metric_names=["L2NormMetric"],
fidelity_features=[2],
)
self.assertTrue(model.use_input_warping)
self.assertTrue(hasattr(model.model, "input_transform"))
self.assertIsInstance(model.model.input_transform, Warp)
# test loocv pseudo likelihood
self.assertFalse(model.use_loocv_pseudo_likelihood)
model = KnowledgeGradient(use_loocv_pseudo_likelihood=True)
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
task_features=[],
feature_names=self.feature_names,
metric_names=["L2NormMetric"],
fidelity_features=[2],
)
self.assertTrue(model.use_loocv_pseudo_likelihood)
def test_KnowledgeGradient_multifidelity(self):
model = KnowledgeGradient()
model.fit(
Xs=self.Xs,
Ys=self.Ys,
Yvars=self.Yvars,
bounds=self.bounds,
task_features=[],
feature_names=self.feature_names,
metric_names=[],
fidelity_features=[-1],
)
# Check best point selection
X_dummy = torch.tensor([1.0, 2.0])
acq_dummy = torch.tensor(0.0)
with mock.patch(self.optimize_acqf,
return_value=(X_dummy,
acq_dummy)) as mock_optimize_acqf:
xbest = model.best_point(
bounds=self.bounds,
objective_weights=self.objective_weights,
target_fidelities={2: 1.0},
)
self.assertTrue(torch.equal(xbest, torch.tensor([1.0, 2.0, 1.0])))
mock_optimize_acqf.assert_called_once()
# check error whenf no target fidelities are specified
with self.assertRaises(RuntimeError):
model.best_point(bounds=self.bounds,
objective_weights=self.objective_weights)
# check generation
n = 2
X_dummy = torch.zeros(12, 1, 2, dtype=self.dtype, device=self.device)
X_dummy2 = torch.zeros(1, n, 3, dtype=self.dtype, device=self.device)
acq_dummy = torch.tensor(0.0, dtype=self.dtype, device=self.device)
dummy1 = (X_dummy, acq_dummy)
dummy2 = (X_dummy2, acq_dummy)
with mock.patch(self.optimize_acqf,
side_effect=[dummy1, dummy2, dummy1,
dummy2]) as mock_optimize_acqf:
Xgen, wgen, _ = model.gen(
n=n,
bounds=self.bounds,
objective_weights=self.objective_weights,
outcome_constraints=None,
linear_constraints=None,
model_gen_options={
"acquisition_function_kwargs": self.acq_options,
"optimizer_kwargs": self.optimizer_options,
},
target_fidelities={2: 1.0},
)
self.assertTrue(torch.equal(Xgen, X_dummy2.cpu()))
self.assertTrue(torch.equal(wgen, torch.ones(n, dtype=self.dtype)))
mock_optimize_acqf.assert_called(
) # called twice, once for best_point
# test error message
linear_constraints = (
torch.tensor([[0.0, 0.0, 0.0], [0.0, 1.0, 0.0]]),
torch.tensor([[0.5], [1.0]]),
)
with self.assertRaises(UnsupportedError):
xbest = model.best_point(
bounds=self.bounds,
linear_constraints=linear_constraints,
objective_weights=self.objective_weights,
target_fidelities={2: 1.0},
)