Remove Models.MOO (facebook#3030)

Summary: Pull Request resolved: facebook#3030 Deprecates `Models.MOO` and points it to `Models.BOTORCH_MODULAR` for backwards compatibility. Reviewed By: Balandat Differential Revision: D65550955 fbshipit-source-id: 4b806ffd48992f965267cdcb8798ee6f5de6cd03
danielcohenlive · Nov 7, 2024 · 4047204 · 4047204
1 parent 6831902
commit 4047204
Show file tree

Hide file tree

Showing 5 changed files with 18 additions and 23 deletions.
diff --git a/ax/modelbridge/registry.py b/ax/modelbridge/registry.py
@@ -188,12 +188,6 @@ class ModelSetup(NamedTuple):
         model_class=UniformGenerator,
         transforms=Cont_X_trans,
     ),
-    "MOO": ModelSetup(
-        bridge_class=TorchModelBridge,
-        model_class=MultiObjectiveBotorchModel,
-        transforms=Cont_X_trans + Y_trans,
-        standard_bridge_kwargs=STANDARD_TORCH_BRIDGE_KWARGS,
-    ),
     "ST_MTGP_LEGACY": ModelSetup(
         bridge_class=TorchModelBridge,
         model_class=BotorchModel,
@@ -433,13 +427,19 @@ class Models(ModelRegistryBase):
     BOTORCH_MODULAR = "BoTorch"
     EMPIRICAL_BAYES_THOMPSON = "EB"
     UNIFORM = "Uniform"
-    MOO = "MOO"
     ST_MTGP_LEGACY = "ST_MTGP_LEGACY"
     ST_MTGP = "ST_MTGP"
     BO_MIXED = "BO_MIXED"
     ST_MTGP_NEHVI = "ST_MTGP_NEHVI"
     CONTEXT_SACBO = "Contextual_SACBO"
 
+    @classmethod
+    @property
+    def MOO(cls) -> Models:
+        return _deprecated_model_with_warning(
+            old_model_str="MOO", new_model=cls.BOTORCH_MODULAR
+        )
+
     @classmethod
     @property
     def GPEI(cls) -> Models:

diff --git a/ax/modelbridge/tests/test_registry.py b/ax/modelbridge/tests/test_registry.py
@@ -505,6 +505,7 @@ def test_deprecated_models(self) -> None:
         same check in a couple different ways.
         """
         for old_model_str, new_model in [
+            ("MOO", Models.BOTORCH_MODULAR),
             ("GPEI", Models.BOTORCH_MODULAR),
             ("FULLYBAYESIAN", Models.SAASBO),
             ("FULLYBAYESIANMOO", Models.SAASBO),

diff --git a/ax/service/tests/test_ax_client.py b/ax/service/tests/test_ax_client.py
@@ -220,7 +220,7 @@ def get_client_with_simple_discrete_moo_problem(
     gs = GenerationStrategy(
         steps=[
             GenerationStep(model=Models.SOBOL, num_trials=3),
-            GenerationStep(model=Models.MOO, num_trials=-1),
+            GenerationStep(model=Models.BOTORCH_MODULAR, num_trials=-1),
         ]
     )
 

diff --git a/ax/service/tests/test_report_utils.py b/ax/service/tests/test_report_utils.py
@@ -442,29 +442,21 @@ def test_get_standard_plots_moo(self) -> None:
         # https://bugs.python.org/issue41943 for more information.
         with self.assertLogs(logger="ax", level=INFO) as log:
             plots = get_standard_plots(
-                experiment=exp, model=Models.MOO(experiment=exp, data=exp.fetch_data())
+                experiment=exp,
+                model=Models.BOTORCH_MODULAR(experiment=exp, data=exp.fetch_data()),
             )
-            self.assertEqual(len(log.output), 5)
+            self.assertEqual(len(log.output), 3)
             self.assertIn(
                 "Pareto plotting not supported for experiments with relative objective "
                 "thresholds.",
                 log.output[0],
             )
-            self.assertIn(
-                "Failed to compute signed global feature sensitivities",
-                log.output[1],
-            )
-            self.assertIn(
-                "Failed to compute unsigned feature sensitivities:",
-                log.output[2],
-            )
-            created_plots_logs = set(log.output[2:])
             for metric_suffix in ("a", "b"):
                 expected_msg = (
                     "Created contour plots for metric branin_"
                     f"{metric_suffix} and parameters ['x2', 'x1']"
                 )
-                self.assertTrue(any(expected_msg in msg for msg in created_plots_logs))
+                self.assertTrue(any(expected_msg in msg for msg in log.output[1:]))
         self.assertEqual(len(plots), 6)
 
     @mock_botorch_optimize
@@ -489,7 +481,8 @@ def test_get_standard_plots_moo_relative_constraints(self) -> None:
         )._objective_thresholds:
             ot.relative = False
         plots = get_standard_plots(
-            experiment=exp, model=Models.MOO(experiment=exp, data=exp.fetch_data())
+            experiment=exp,
+            model=Models.BOTORCH_MODULAR(experiment=exp, data=exp.fetch_data()),
         )
         self.assertEqual(len(plots), 8)
 
@@ -500,7 +493,8 @@ def test_get_standard_plots_moo_no_objective_thresholds(self) -> None:
         exp.optimization_config.objective.objectives[1].minimize = True
         exp.trials[0].run()
         plots = get_standard_plots(
-            experiment=exp, model=Models.MOO(experiment=exp, data=exp.fetch_data())
+            experiment=exp,
+            model=Models.BOTORCH_MODULAR(experiment=exp, data=exp.fetch_data()),
         )
         self.assertEqual(len(plots), 8)
 

diff --git a/ax/service/utils/best_point.py b/ax/service/utils/best_point.py
@@ -558,7 +558,7 @@ def get_pareto_optimal_parameters(
     if is_moo_modelbridge:
         generation_strategy._fit_current_model(data=None)
     else:
-        modelbridge = Models.MOO(
+        modelbridge = Models.BOTORCH_MODULAR(
             experiment=experiment,
             data=checked_cast(
                 Data, experiment.lookup_data(trial_indices=trial_indices)