Remove_outcome_transform=Standardize for SingleTaskGP [documentation] (#2966)

Summary:
Pull Request resolved: #2966

Default behavior of SingleTaskGP is to standardize outcomes, changed examples in various documentation to reflect this.

Reviewed By: saitcakmak

Differential Revision: D80029965

fbshipit-source-id: 3a651aa3d21088a90a97c46f5f6e393b50c53b98

Author: Ethan Che

README.md

@@ -136,7 +136,7 @@ For more details see our [Documentation](https://botorch.org/docs/introduction)
   ```python
   import torch
   from botorch.models import SingleTaskGP
-  from botorch.models.transforms import Normalize, Standardize
+  from botorch.models.transforms import Normalize
   from botorch.fit import fit_gpytorch_mll
   from gpytorch.mlls import ExactMarginalLogLikelihood
 
@@ -150,7 +150,6 @@ For more details see our [Documentation](https://botorch.org/docs/introduction)
       train_X=train_X,
       train_Y=Y,
       input_transform=Normalize(d=2),
-      outcome_transform=Standardize(m=1),
   )
   mll = ExactMarginalLogLikelihood(gp.likelihood, gp)
   fit_gpytorch_mll(mll)

botorch/models/gp_regression.py

@@ -90,30 +90,20 @@ class SingleTaskGP(BatchedMultiOutputGPyTorchModel, ExactGP, FantasizeMixin):
 
         >>> import torch
         >>> from botorch.models.gp_regression import SingleTaskGP
-        >>> from botorch.models.transforms.outcome import Standardize
         >>>
         >>> train_X = torch.rand(20, 2, dtype=torch.float64)
         >>> train_Y = torch.sin(train_X).sum(dim=1, keepdim=True)
-        >>> outcome_transform = Standardize(m=1)
-        >>> inferred_noise_model = SingleTaskGP(
-        ...     train_X, train_Y, outcome_transform=outcome_transform,
-        ... )
+        >>> inferred_noise_model = SingleTaskGP(train_X, train_Y)
 
         Model with a known observation variance of 0.2:
 
         >>> train_Yvar = torch.full_like(train_Y, 0.2)
-        >>> observed_noise_model = SingleTaskGP(
-        ...     train_X, train_Y, train_Yvar,
-        ...     outcome_transform=outcome_transform,
-        ... )
+        >>> observed_noise_model = SingleTaskGP(train_X, train_Y, train_Yvar)
 
         With noise-free observations:
 
         >>> train_Yvar = torch.full_like(train_Y, 1e-6)
-        >>> noise_free_model = SingleTaskGP(
-        ...     train_X, train_Y, train_Yvar,
-        ...     outcome_transform=outcome_transform,
-        ... )
+        >>> noise_free_model = SingleTaskGP(train_X, train_Y, train_Yvar)
     """
 
     train_targets: Tensor

docs/getting_started.mdx

@@ -49,7 +49,7 @@ Here's a quick run down of the main components of a Bayesian Optimization loop.
     ```python
     import torch
     from botorch.models import SingleTaskGP
-    from botorch.models.transforms import Normalize, Standardize
+    from botorch.models.transforms import Normalize
     from botorch.fit import fit_gpytorch_mll
     from gpytorch.mlls import ExactMarginalLogLikelihood
 
@@ -62,7 +62,6 @@ Here's a quick run down of the main components of a Bayesian Optimization loop.
         train_X=train_X,
         train_Y=train_Y,
         input_transform=Normalize(d=2),
-        outcome_transform=Standardize(m=1),
     )
     mll = ExactMarginalLogLikelihood(gp.likelihood, gp)
     fit_gpytorch_mll(mll)