Revert initialize acqf method to not add bounds to calls (#490)

Summary:
Pull Request resolved: #490

Acqf_kwargs already gets bounds from config if it needs it so now we work with those instead of the generator's bounds.

Reviewed By: crasanders

Differential Revision: D67532416

fbshipit-source-id: 7fc57d320213dfedf874249da4f225371be40742

aepsych/generators/optimize_acqf_generator.py

@@ -82,33 +82,26 @@ def _instantiate_acquisition_fn(self, model: ModelProtocol) -> AcquisitionFuncti
         Returns:
             AcquisitionFunction: Configured acquisition function.
         """
-        if (
-            "lb" in inspect.signature(self.acqf).parameters
-            and "ub" in inspect.signature(self.acqf).parameters
-        ):
-            if self.acqf == AnalyticExpectedUtilityOfBestOption:
-                return self.acqf(pref_model=model, lb=self.lb, ub=self.ub)
+        if self.acqf == AnalyticExpectedUtilityOfBestOption:
+            return self.acqf(pref_model=model)
 
-            self.lb = self.lb.to(model.device)
-            self.ub = self.ub.to(model.device)
-            if self.acqf in self.baseline_requiring_acqfs:
-                return self.acqf(
-                    model,
-                    model.train_inputs[0],
-                    lb=self.lb,
-                    ub=self.ub,
-                    **self.acqf_kwargs,
-                )
+        if hasattr(model, "device"):
+            if "lb" in self.acqf_kwargs:
+                if not isinstance(self.acqf_kwargs["lb"], torch.Tensor):
+                    self.acqf_kwargs["lb"] = torch.tensor(self.acqf_kwargs["lb"])
 
-            return self.acqf(model=model, lb=self.lb, ub=self.ub, **self.acqf_kwargs)
+                self.acqf_kwargs["lb"] = self.acqf_kwargs["lb"].to(model.device)
 
-        if self.acqf == AnalyticExpectedUtilityOfBestOption:
-            return self.acqf(pref_model=model)
+            if "ub" in self.acqf_kwargs:
+                if not isinstance(self.acqf_kwargs["ub"], torch.Tensor):
+                    self.acqf_kwargs["ub"] = torch.tensor(self.acqf_kwargs["ub"])
+
+                self.acqf_kwargs["ub"] = self.acqf_kwargs["ub"].to(model.device)
 
         if self.acqf in self.baseline_requiring_acqfs:
             return self.acqf(model, model.train_inputs[0], **self.acqf_kwargs)
-
-        return self.acqf(model=model, **self.acqf_kwargs)
+        else:
+            return self.acqf(model=model, **self.acqf_kwargs)
 
     def gen(self, num_points: int, model: ModelProtocol, **gen_options) -> torch.Tensor:
         """Query next point(s) to run by optimizing the acquisition function.

tests/models/test_semi_p.py

@@ -141,6 +141,8 @@ def test_analytic_lookahead_generation(self):
                 "target": 0.75,
                 "query_set_size": 100,
                 "Xq": make_scaled_sobol(self.lb, self.ub, 100),
+                "lb": self.lb,
+                "ub": self.ub,
             },
             max_gen_time=0.2,
             lb=self.lb,

tests_gpu/generators/test_optimize_acqf_generator.py

@@ -6,6 +6,7 @@
 # LICENSE file in the root directory of this source tree.
 
 import unittest
+from inspect import signature
 
 import torch
 from aepsych.acquisition import (
@@ -51,9 +52,14 @@ class TestOptimizeAcqfGenerator(unittest.TestCase):
     def test_gpu_smoketest(self, acqf, acqf_kwargs):
         lb = torch.tensor([0.0])
         ub = torch.tensor([1.0])
-        bounds = torch.stack([lb, ub])
         inducing_size = 10
 
+        acqf_args_expected = list(signature(acqf).parameters.keys())
+        if "lb" in acqf_args_expected:
+            acqf_kwargs = acqf_kwargs.copy()
+            acqf_kwargs["lb"] = lb
+            acqf_kwargs["ub"] = ub
+
         model = GPClassificationModel(
             dim=1,
             inducing_size=inducing_size,