Don't raise specious warning about input transforms with approximate …

…GPs (#1826)

Summary:
## Motivation

The BoTorch base `Model` class warns if an input transform has been provided, the `eval` method is called, and the object has no `train_inputs` attribute. This is not appropriate for `ApproximateGPyTorchModel`s; see #1824 . This PR gives `ApproximateGPyTorchModel` the `train` and `eval` modes from `torch.nn.Module`, which is the same as the methods it had been inheriting from `Model` but without the irrelevant input transform logic.

A nicer fix would be to remove the input transform logic from `Model` and have it only in subclasses that it applies to, so that subclasses like `ApproximateGPyTorchModel` would not need to do anything special to avoid inheriting that.

I think this all applies to `EnsembleModel`s as well as `ApproximateGPyTorchModel`s --looking into this now.

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

Pull Request resolved: #1826

Test Plan: Existing units for `ApproximateGPyTorchModel` look good.

Reviewed By: Balandat

Differential Revision: D45782048

Pulled By: esantorella

fbshipit-source-id: 2091956a5a0cb6680f4c7292c0951f9079975ffb

botorch/models/approximate_gp.py

@@ -32,7 +32,7 @@
 import copy
 import warnings
 
-from typing import Optional, Type, Union
+from typing import Optional, Type, TypeVar, Union
 
 import torch
 from botorch.models.gpytorch import GPyTorchModel
@@ -64,11 +64,15 @@
     VariationalStrategy,
 )
 from torch import Tensor
+from torch.nn import Module
 
 
 MIN_INFERRED_NOISE_LEVEL = 1e-4
 
 
+TApproxModel = TypeVar("TApproxModel", bound="ApproximateGPyTorchModel")
+
+
 class ApproximateGPyTorchModel(GPyTorchModel):
     r"""
     Botorch wrapper class for various (variational) approximate GP models in
@@ -120,6 +124,19 @@ def __init__(
     def num_outputs(self):
         return self._desired_num_outputs
 
+    def eval(self: TApproxModel) -> TApproxModel:
+        r"""Puts the model in `eval` mode."""
+        return Module.eval(self)
+
+    def train(self: TApproxModel, mode: bool = True) -> TApproxModel:
+        r"""Put the model in `train` mode.
+
+        Args:
+            mode: A boolean denoting whether to put in `train` or `eval` mode.
+                If `False`, model is put in `eval` mode.
+        """
+        return Module.train(self, mode=mode)
+
     def posterior(
         self, X, output_indices=None, observation_noise=False, *args, **kwargs
     ) -> GPyTorchPosterior:

test/models/test_approximate_gp.py

@@ -7,6 +7,7 @@
 import itertools
 
 import torch
+from botorch.fit import fit_gpytorch_mll
 from botorch.models.approximate_gp import (
     _SingleTaskVariationalGP,
     ApproximateGPyTorchModel,
@@ -307,3 +308,19 @@ def test_custom_inducing_point_init(self):
         self.assertEqual(model_2_inducing.shape, (5, 1))
         self.assertAllClose(model_1_inducing, model_2_inducing)
         self.assertFalse(model_1_inducing[0, 0] == model_3_inducing[0, 0])
+
+    def test_input_transform(self) -> None:
+        train_X = torch.linspace(1, 3, 10, dtype=torch.double)[:, None]
+        y = -3 * train_X + 5
+
+        for input_transform in [None, Normalize(1)]:
+            with self.subTest(input_transform=input_transform):
+                model = SingleTaskVariationalGP(
+                    train_X=train_X, train_Y=y, input_transform=input_transform
+                )
+                mll = VariationalELBO(
+                    model.likelihood, model.model, num_data=train_X.shape[-2]
+                )
+                fit_gpytorch_mll(mll)
+                post = model.posterior(torch.tensor([train_X.mean()]))
+                self.assertAllClose(post.mean[0][0], y.mean(), atol=1e-4)