Update GenerationNode.model_to_gen_from_name

Summary: Updates the behavior to match the docstring. Previously, this would call `model_spec_to_gen_from`, which could lead to fitting multiple models if the node had multiple model specs. The return type was always `str` rather than `Optional[str]` suggested by the type hints.

Differential Revision: D56735619

ax/modelbridge/generation_node.py

@@ -143,7 +143,10 @@ def model_to_gen_from_name(self) -> Optional[str]:
         """Returns the name of the model that will be used for gen, if there is one.
         Otherwise, returns None.
         """
-        return self.model_spec_to_gen_from.model_key
+        if self._model_spec_to_gen_from is not None:
+            return self._model_spec_to_gen_from.model_key
+        else:
+            return None
 
     @property
     def generation_strategy(self) -> modelbridge.generation_strategy.GenerationStrategy:

ax/modelbridge/tests/test_generation_node.py

@@ -113,6 +113,7 @@ def test_properties(self) -> None:
                 ),
             ],
         )
+        self.assertIsNone(node.model_to_gen_from_name)
         dat = self.branin_experiment.lookup_data()
         node.fit(
             experiment=self.branin_experiment,
@@ -124,6 +125,7 @@ def test_properties(self) -> None:
         self.assertEqual(
             node.model_spec_to_gen_from.model_kwargs, node.model_specs[0].model_kwargs
         )
+        self.assertEqual(node.model_to_gen_from_name, "GPEI")
         self.assertEqual(
             node.model_spec_to_gen_from.model_gen_kwargs,
             node.model_specs[0].model_gen_kwargs,