Patch codecov for SaasFullyBayesianSingleTaskGP & FullyBayesianPosterior

Summary: --

Reviewed By: Balandat, SebastianAment

Differential Revision: D54278602

fbshipit-source-id: 259fc73e03a68a13507b1574c50ca21fca5d86b2

test/models/test_fully_bayesian.py

@@ -50,7 +50,11 @@
  SaasPyroModel,
 )
 from botorch.models.transforms import Normalize, Standardize
-from botorch.posteriors.fully_bayesian import batched_bisect, GaussianMixturePosterior
+from botorch.posteriors.fully_bayesian import (
+ batched_bisect,
+ FullyBayesianPosterior,
+ GaussianMixturePosterior,
+)
 from botorch.sampling.get_sampler import get_sampler
 from botorch.utils.datasets import SupervisedDataset
 from botorch.utils.multi_objective.box_decompositions.non_dominated import (
@@ -752,6 +756,15 @@ def test_condition_on_observation(self):
  torch.Size([num_models, num_train + num_cond, num_dims]),
  )
 
+ # With batch size only on Y.
+ cond_model = model.condition_on_observations(
+ cond_X_nobatch, cond_Y, noise=cond_Yvar
+ )
+ self.assertEqual(
+ cond_model.train_inputs[0].shape,
+ torch.Size([num_models, num_train + num_cond, num_dims]),
+ )
+
  # test repeated conditining
  repeat_cond_X = cond_X + 5
  repeat_cond_model = cond_model.condition_on_observations(
@@ -815,6 +828,17 @@ def f(x):
  dist.cdf(x), q * torch.ones(1, 5, 1, **tkwargs), atol=1e-4
  )
 
+ def test_deprecated_posterior(self) -> None:
+ mean = torch.randn(1, 5)
+ variance = torch.rand(1, 5)
+ covar = torch.diag_embed(variance)
+ mvn = MultivariateNormal(mean, to_linear_operator(covar))
+ with self.assertWarnsRegex(
+ DeprecationWarning, "`FullyBayesianPosterior` is marked for deprecation"
+ ):
+ posterior = FullyBayesianPosterior(distribution=mvn)
+ self.assertIsInstance(posterior, GaussianMixturePosterior)
+
 
 class TestPyroCatchNumericalErrors(BotorchTestCase):
  def tearDown(self):