Fix build issues with sphinx 3.0.0 (pytorch#415)

Summary:
sphinx 3.0.0 was just released with some modified behavior, which caused some issues with repeated defs. This fixes these issues and makes the spinx build pass on the current release.

This also spews a large number of additional deprecation warnings, will try to address these in a different PR. .
Pull Request resolved: pytorch#415

Reviewed By: sdsingh

Differential Revision: D20862807

Pulled By: Balandat

fbshipit-source-id: 500a0e3d00ea0c6e16410481c11155837e077be3

Author: Balandat

botorch/sampling/samplers.py

@@ -29,7 +29,6 @@ class MCSampler(Module, ABC):
     Subclasses must implement the `_construct_base_samples` method.
 
     Attributes:
-        sample_shape: The shape of each sample.
         resample: If `True`, re-draw samples in each `forward` evaluation -
             this results in stochastic acquisition functions (and thus should
             not be used with deterministic optimization algorithms).

sphinx/source/sampling.rst

@@ -7,12 +7,6 @@ botorch.sampling
 .. automodule:: botorch.sampling
 
 
-Monte-Carlo Sampler API
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-.. currentmodule:: botorch.sampling.samplers
-.. autoclass:: MCSampler
-    :members:
-
 Monte-Carlo Samplers
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 .. automodule:: botorch.sampling.samplers