Merge branch 'main' into np_regression

Author: eibarolle

botorch/acquisition/input_constructors.py

@@ -1758,7 +1758,7 @@ def optimize_objective(
             columns=list(fixed_features.keys()),
             values=list(fixed_features.values()),
         )
-        free_feature_dims = list(range(len(bounds)) - fixed_features.keys())
+        free_feature_dims = list(range(bounds.shape[1]) - fixed_features.keys())
         free_feature_bounds = bounds[:, free_feature_dims]  # (2, d' <= d)
     else:
         free_feature_bounds = bounds
@@ -1775,18 +1775,21 @@ def optimize_objective(
             rhs = -b[i, 0]
             inequality_constraints.append((indices, coefficients, rhs))
 
+    options = {
+        "batch_limit": optimizer_options.get("batch_limit", 8),
+        "maxiter": optimizer_options.get("maxiter", 200),
+        "nonnegative": optimizer_options.get("nonnegative", False),
+    }
+    if "method" in optimizer_options:
+        options["method"] = optimizer_options.pop("method")
+
     return optimize_acqf(
         acq_function=acq_function,
         bounds=free_feature_bounds,
         q=q,
         num_restarts=optimizer_options.get("num_restarts", 60),
         raw_samples=optimizer_options.get("raw_samples", 1024),
-        options={
-            "batch_limit": optimizer_options.get("batch_limit", 8),
-            "maxiter": optimizer_options.get("maxiter", 200),
-            "nonnegative": optimizer_options.get("nonnegative", False),
-            "method": optimizer_options.get("method", "L-BFGS-B"),
-        },
+        options=options,
         inequality_constraints=inequality_constraints,
         fixed_features=None,  # handled inside the acquisition function
         post_processing_func=post_processing_func,

botorch/acquisition/joint_entropy_search.py

@@ -29,7 +29,7 @@
 from botorch import settings
 from botorch.acquisition.acquisition import AcquisitionFunction, MCSamplerMixin
 from botorch.acquisition.objective import PosteriorTransform
-from botorch.models.fully_bayesian import SaasFullyBayesianSingleTaskGP
+from botorch.models.fully_bayesian import FullyBayesianSingleTaskGP
 from botorch.models.model import Model
 from botorch.models.utils import check_no_nans, fantasize as fantasize_flag
 from botorch.models.utils.gpytorch_modules import MIN_INFERRED_NOISE_LEVEL
@@ -124,7 +124,7 @@ def __init__(
         # and the optimal outputs have shapes num_optima x [num_models if FB] x 1 x 1
         # The third dimension equaling 1 is required to get one optimum per model,
         # which raises a BotorchTensorDimensionWarning.
-        if isinstance(model, SaasFullyBayesianSingleTaskGP):
+        if isinstance(model, FullyBayesianSingleTaskGP):
             raise NotImplementedError(FULLY_BAYESIAN_ERROR_MSG)
         with warnings.catch_warnings():
             warnings.filterwarnings("ignore")