prevent warning (meta-pytorch#3078)

Summary:
<!--
Thank you for sending the PR! We appreciate you spending the time to make BoTorch better.

Help us understand your motivation by explaining why you decided to make this change.

You can learn more about contributing to BoTorch here: https://github.com/meta-pytorch/botorch/blob/main/CONTRIBUTING.md
-->

## Motivation

Fixes the issue described in meta-pytorch#3071.

### Have you read the [Contributing Guidelines on pull requests](https://github.com/meta-pytorch/botorch/blob/main/CONTRIBUTING.md#pull-requests)?

Yes.

Pull Request resolved: meta-pytorch#3078

Test Plan: Unit tests.

Reviewed By: hvarfner

Differential Revision: D86792344

Pulled By: Balandat

fbshipit-source-id: 6d5e4886b81e304efbea814847b3fe03674c799c

Author: jduerholt

botorch/models/deterministic.py

@@ -40,6 +40,12 @@
 class DeterministicModel(EnsembleModel):
     """Abstract base class for deterministic models."""
 
+    def _set_transformed_inputs(self):
+        """Overwrites the parent method to prevent raise of
+        warning "Could not update `train_inputs` with transformed inputs."
+        """
+        return None
+
     @abstractmethod
     def forward(self, X: Tensor) -> Tensor:
         r"""Compute the (deterministic) model output at X.

test/models/test_deterministic.py

@@ -4,7 +4,6 @@
 # This source code is licensed under the MIT license found in the
 # LICENSE file in the root directory of this source tree.
 
-import warnings
 
 import torch
 from botorch.acquisition.objective import ScalarizedPosteriorTransform
@@ -149,15 +148,11 @@ def test_with_transforms(self):
         # check that the posterior output agrees with the manually transformed one
         test_X = torch.rand(3, dim)
         expected_Y, _ = octf.untransform(model.forward(intf(test_X)))
-        with warnings.catch_warnings(record=True) as ws:
-            posterior = model.posterior(test_X)
-            msg = "does not have a `train_inputs` attribute"
-            self.assertTrue(any(msg in str(w.message) for w in ws))
+        posterior = model.posterior(test_X)
         self.assertAllClose(expected_Y, posterior.mean)
-        # check that model.train/eval works and raises the warning
+        # check that model.train/eval works
         model.train()
-        with self.assertWarns(RuntimeWarning):
-            model.eval()
+        model.eval()
 
     def test_posterior_transform(self):
         def f(X):