Delete references to deprecated DeterministicPosterior and Determinis…

…ticSampler (pytorch#2391)

Summary:
## Motivation

Delete references to deprecated DeterministicPosterior and DeterministicSampler, replacing `DeterministicPosterior` with `EnsemblePosterior` where appropriate. These were deprecated before 0.9.0, so now that we are past 0.11.0, they can be reaped.

Pull Request resolved: pytorch#2391

Test Plan:
Replaced `DeterministicPosterior` with `EnsemblePosterior` in tests

## Related PRs

pytorch#1636

Reviewed By: Balandat

Differential Revision: D59057165

Pulled By: esantorella

fbshipit-source-id: 70a8405d0e75d414a685192808e8c0b18a6aca92

botorch/posteriors/__init__.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.
 
-from botorch.posteriors.deterministic import DeterministicPosterior
 from botorch.posteriors.fully_bayesian import (
     FullyBayesianPosterior,
     GaussianMixturePosterior,
@@ -18,7 +17,6 @@
 from botorch.posteriors.transformed import TransformedPosterior
 
 __all__ = [
-    "DeterministicPosterior",
     "GaussianMixturePosterior",
     "FullyBayesianPosterior",
     "GPyTorchPosterior",

botorch/posteriors/deterministic.py

@@ -3,89 +3,3 @@
 #
 # This source code is licensed under the MIT license found in the
 # LICENSE file in the root directory of this source tree.
-
-r"""
-Deterministic (degenerate) posteriors. Used in conjunction with deterministic
-models.
-"""
-
-from __future__ import annotations
-
-from typing import Optional
-from warnings import warn
-
-import torch
-from botorch.posteriors.posterior import Posterior
-from torch import Tensor
-
-
-class DeterministicPosterior(Posterior):
-    r"""Deterministic posterior.
-
-    [DEPRECATED] Use `EnsemblePosterior` instead.
-    """
-
-    def __init__(self, values: Tensor) -> None:
-        r"""
-        Args:
-            values: Values of the samples produced by this posterior.
-        """
-        warn(
-            "`DeterministicPosterior` is marked for deprecation, consider using "
-            "`EnsemblePosterior`.",
-            DeprecationWarning,
-        )
-        self.values = values
-
-    @property
-    def device(self) -> torch.device:
-        r"""The torch device of the posterior."""
-        return self.values.device
-
-    @property
-    def dtype(self) -> torch.dtype:
-        r"""The torch dtype of the posterior."""
-        return self.values.dtype
-
-    def _extended_shape(
-        self, sample_shape: torch.Size = torch.Size()  # noqa: B008
-    ) -> torch.Size:
-        r"""Returns the shape of the samples produced by the posterior with
-        the given `sample_shape`.
-        """
-        return sample_shape + self.values.shape
-
-    @property
-    def mean(self) -> Tensor:
-        r"""The mean of the posterior as a `(b) x n x m`-dim Tensor."""
-        return self.values
-
-    @property
-    def variance(self) -> Tensor:
-        r"""The variance of the posterior as a `(b) x n x m`-dim Tensor.
-
-        As this is a deterministic posterior, this is a tensor of zeros.
-        """
-        return torch.zeros_like(self.values)
-
-    def rsample(
-        self,
-        sample_shape: Optional[torch.Size] = None,
-    ) -> Tensor:
-        r"""Sample from the posterior (with gradients).
-
-        For the deterministic posterior, this just returns the values expanded
-        to the requested shape.
-
-        Args:
-            sample_shape: A `torch.Size` object specifying the sample shape. To
-                draw `n` samples, set to `torch.Size([n])`. To draw `b` batches
-                of `n` samples each, set to `torch.Size([b, n])`.
-
-        Returns:
-            Samples from the posterior, a tensor of shape
-            `self._extended_shape(sample_shape=sample_shape)`.
-        """
-        if sample_shape is None:
-            sample_shape = torch.Size([1])
-        return self.values.expand(self._extended_shape(sample_shape))

botorch/sampling/__init__.py

@@ -5,7 +5,6 @@
 # LICENSE file in the root directory of this source tree.
 
 from botorch.sampling.base import MCSampler
-from botorch.sampling.deterministic import DeterministicSampler
 from botorch.sampling.get_sampler import get_sampler
 from botorch.sampling.list_sampler import ListSampler
 from botorch.sampling.normal import IIDNormalSampler, SobolQMCNormalSampler
@@ -20,7 +19,6 @@
 
 
 __all__ = [
-    "DeterministicSampler",
     "ForkedRNGSampler",
     "get_sampler",
     "IIDNormalSampler",

botorch/sampling/deterministic.py

@@ -3,35 +3,3 @@
 #
 # This source code is licensed under the MIT license found in the
 # LICENSE file in the root directory of this source tree.
-
-r"""
-A dummy sampler for use with deterministic models.
-"""
-
-from __future__ import annotations
-
-from botorch.posteriors.deterministic import DeterministicPosterior
-from botorch.sampling.stochastic_samplers import StochasticSampler
-
-
-class DeterministicSampler(StochasticSampler):
-    r"""A sampler that simply calls `posterior.rsample`, intended to be used with
-    `DeterministicModel` & `DeterministicPosterior`.
-
-    [DEPRECATED] - Use `IndexSampler` in conjunction with `EnsemblePosterior`
-    instead of `DeterministicSampler` with `DeterministicPosterior`.
-
-    This is effectively signals that `StochasticSampler` is safe to use with
-    deterministic models since their output is deterministic by definition.
-    """
-
-    def _update_base_samples(
-        self, posterior: DeterministicPosterior, base_sampler: DeterministicSampler
-    ) -> None:
-        r"""This is a no-op since there are no base samples to update.
-
-        Args:
-            posterior: The posterior for which the base samples are constructed.
-            base_sampler: The base sampler to retrieve the base samples from.
-        """
-        return

botorch/sampling/get_sampler.py

@@ -9,15 +9,13 @@
 
 import torch
 from botorch.logging import logger
-from botorch.posteriors.deterministic import DeterministicPosterior
 from botorch.posteriors.ensemble import EnsemblePosterior
 from botorch.posteriors.gpytorch import GPyTorchPosterior
 from botorch.posteriors.posterior import Posterior
 from botorch.posteriors.posterior_list import PosteriorList
 from botorch.posteriors.torch import TorchPosterior
 from botorch.posteriors.transformed import TransformedPosterior
 from botorch.sampling.base import MCSampler
-from botorch.sampling.deterministic import DeterministicSampler
 from botorch.sampling.index_sampler import IndexSampler
 from botorch.sampling.list_sampler import ListSampler
 from botorch.sampling.normal import (
@@ -119,16 +117,6 @@ def _get_sampler_list(
     return ListSampler(*samplers)
 
 
-@GetSampler.register(DeterministicPosterior)
-def _get_sampler_deterministic(
-    posterior: DeterministicPosterior,
-    sample_shape: torch.Size,
-    seed: Optional[int] = None,
-) -> MCSampler:
-    r"""Get the dummy `DeterministicSampler` for the `DeterministicPosterior`."""
-    return DeterministicSampler(sample_shape=sample_shape, seed=seed)
-
-
 @GetSampler.register(EnsemblePosterior)
 def _get_sampler_ensemble(
     posterior: EnsemblePosterior,

test/models/test_model.py

@@ -10,7 +10,7 @@
 from botorch.exceptions.errors import InputDataError
 from botorch.models.deterministic import GenericDeterministicModel
 from botorch.models.model import Model, ModelDict, ModelList
-from botorch.posteriors.deterministic import DeterministicPosterior
+from botorch.posteriors.ensemble import EnsemblePosterior
 from botorch.posteriors.posterior_list import PosteriorList
 from botorch.utils.datasets import SupervisedDataset
 from botorch.utils.testing import BotorchTestCase, MockModel, MockPosterior
@@ -35,7 +35,10 @@ def evaluate(self, Y):
 
     def forward(self, posterior):
         return PosteriorList(
-            *[DeterministicPosterior(2 * p.mean + 1) for p in posterior.posteriors]
+            *[
+                EnsemblePosterior(2 * p.mean.unsqueeze(0) + 1)
+                for p in posterior.posteriors
+            ]
         )
 
 

test/posteriors/test_deterministic.py

@@ -3,37 +3,3 @@
 #
 # This source code is licensed under the MIT license found in the
 # LICENSE file in the root directory of this source tree.
-
-import itertools
-
-from warnings import catch_warnings
-
-import torch
-from botorch.posteriors.deterministic import DeterministicPosterior
-from botorch.utils.testing import BotorchTestCase
-
-
-class TestDeterministicPosterior(BotorchTestCase):
-    def test_DeterministicPosterior(self):
-        for shape, dtype in itertools.product(
-            ((3, 2), (2, 3, 1)), (torch.float, torch.double)
-        ):
-            values = torch.randn(*shape, device=self.device, dtype=dtype)
-            p = DeterministicPosterior(values)
-            with catch_warnings(record=True) as ws:
-                p = DeterministicPosterior(values)
-                self.assertTrue(
-                    any("marked for deprecation" in str(w.message) for w in ws)
-                )
-            self.assertEqual(p.device.type, self.device.type)
-            self.assertEqual(p.dtype, dtype)
-            self.assertEqual(p._extended_shape(), values.shape)
-            with self.assertRaises(NotImplementedError):
-                p.base_sample_shape
-            self.assertTrue(torch.equal(p.mean, values))
-            self.assertTrue(torch.equal(p.variance, torch.zeros_like(values)))
-            # test sampling
-            samples = p.rsample()
-            self.assertTrue(torch.equal(samples, values.unsqueeze(0)))
-            samples = p.rsample(torch.Size([2]))
-            self.assertTrue(torch.equal(samples, values.expand(2, *values.shape)))

test/posteriors/test_posterior.py

@@ -9,7 +9,7 @@
 
 import torch
 from botorch.posteriors import GPyTorchPosterior, Posterior, PosteriorList
-from botorch.posteriors.deterministic import DeterministicPosterior
+from botorch.posteriors.ensemble import EnsemblePosterior
 from botorch.utils.testing import BotorchTestCase
 from gpytorch.distributions import MultivariateNormal
 from linear_operator.operators import to_linear_operator
@@ -57,7 +57,7 @@ def _make_gpytorch_posterior(self, shape, dtype):
 
     def _make_deterministic_posterior(self, shape, dtype):
         mean = torch.rand(*shape, 1, dtype=dtype, device=self.device)
-        return DeterministicPosterior(values=mean)
+        return EnsemblePosterior(values=mean.unsqueeze(0))
 
     def test_posterior_list(self):
         for dtype, use_deterministic in product(

test/sampling/test_deterministic.py

@@ -3,22 +3,3 @@
 #
 # This source code is licensed under the MIT license found in the
 # LICENSE file in the root directory of this source tree.
-
-import torch
-from botorch.sampling.deterministic import DeterministicSampler
-from botorch.utils.testing import BotorchTestCase, MockPosterior
-
-
-class TestDeterministicSampler(BotorchTestCase):
-    def test_deterministic_sampler(self):
-        # Basic usage.
-        samples = torch.rand(1, 2)
-        posterior = MockPosterior(samples=samples)
-        sampler = DeterministicSampler(sample_shape=torch.Size([2]))
-        self.assertTrue(torch.equal(samples.repeat(2, 1, 1), sampler(posterior)))
-
-        # Test _update_base_samples.
-        sampler._update_base_samples(
-            posterior=posterior,
-            base_sampler=sampler,
-        )

test/sampling/test_get_sampler.py

@@ -5,15 +5,13 @@
 # LICENSE file in the root directory of this source tree.
 
 import torch
-from botorch.posteriors.deterministic import DeterministicPosterior
 from botorch.posteriors.gpytorch import GPyTorchPosterior
 from botorch.posteriors.posterior_list import PosteriorList
 from botorch.posteriors.torch import TorchPosterior
 from botorch.posteriors.transformed import TransformedPosterior
 from botorch.sampling.get_sampler import get_sampler
 from botorch.sampling.list_sampler import ListSampler
 from botorch.sampling.normal import IIDNormalSampler, SobolQMCNormalSampler
-from botorch.sampling.stochastic_samplers import StochasticSampler
 from botorch.utils.testing import BotorchTestCase
 from gpytorch.distributions import MultivariateNormal
 from torch.distributions.gamma import Gamma
@@ -22,40 +20,42 @@
 class TestGetSampler(BotorchTestCase):
     def test_get_sampler(self):
         # Basic usage w/ gpytorch posterior.
-        posterior = GPyTorchPosterior(
+        mvn_posterior = GPyTorchPosterior(
             distribution=MultivariateNormal(torch.rand(2), torch.eye(2))
         )
+        seed = 2
+        n_samples = 10
         sampler = get_sampler(
-            posterior=posterior, sample_shape=torch.Size([10]), seed=2
+            posterior=mvn_posterior, sample_shape=torch.Size([n_samples]), seed=seed
         )
         self.assertIsInstance(sampler, SobolQMCNormalSampler)
-        self.assertEqual(sampler.seed, 2)
-        self.assertEqual(sampler.sample_shape, torch.Size([10]))
+        self.assertEqual(sampler.seed, seed)
+        self.assertEqual(sampler.sample_shape, torch.Size([n_samples]))
 
         # Fallback to IID sampler.
-        posterior = GPyTorchPosterior(
+        big_mvn_posterior = GPyTorchPosterior(
             distribution=MultivariateNormal(torch.rand(22000), torch.eye(22000))
         )
-        sampler = get_sampler(posterior=posterior, sample_shape=torch.Size([10]))
+        sampler = get_sampler(
+            posterior=big_mvn_posterior, sample_shape=torch.Size([n_samples])
+        )
         self.assertIsInstance(sampler, IIDNormalSampler)
-        self.assertEqual(sampler.sample_shape, torch.Size([10]))
+        self.assertEqual(sampler.sample_shape, torch.Size([n_samples]))
 
         # Transformed posterior.
         tf_post = TransformedPosterior(
-            posterior=posterior, sample_transform=lambda X: X
+            posterior=big_mvn_posterior, sample_transform=lambda X: X
         )
-        sampler = get_sampler(posterior=tf_post, sample_shape=torch.Size([10]))
+        sampler = get_sampler(posterior=tf_post, sample_shape=torch.Size([n_samples]))
         self.assertIsInstance(sampler, IIDNormalSampler)
-        self.assertEqual(sampler.sample_shape, torch.Size([10]))
+        self.assertEqual(sampler.sample_shape, torch.Size([n_samples]))
 
-        # PosteriorList with transformed & deterministic.
-        post_list = PosteriorList(
-            tf_post, DeterministicPosterior(values=torch.rand(1, 2))
-        )
+        # PosteriorList with transformed & original
+        post_list = PosteriorList(tf_post, mvn_posterior)
         sampler = get_sampler(posterior=post_list, sample_shape=torch.Size([5]))
         self.assertIsInstance(sampler, ListSampler)
         self.assertIsInstance(sampler.samplers[0], IIDNormalSampler)
-        self.assertIsInstance(sampler.samplers[1], StochasticSampler)
+        self.assertIsInstance(sampler.samplers[1], SobolQMCNormalSampler)
         for s in sampler.samplers:
             self.assertEqual(s.sample_shape, torch.Size([5]))