Provide more informative warning messages in InputDataWarning (face…

…book#2713) Summary: Pull Request resolved: facebook#2713 X-link: pytorch/botorch#2489 Provide more informative warning messages when `InputDataWarning`s are raised to specify whether it pertains to input features or output targets. Updated unit tests accordingly to ensure coverage. Differential Revision: D61797434
ltiao · Aug 28, 2024 · 89d5a34 · 89d5a34
1 parent 5ef2c61
commit 89d5a34
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 3 deletions.
diff --git a/ax/modelbridge/base.py b/ax/modelbridge/base.py
@@ -904,7 +904,7 @@ def cross_validate(
             # users with this warning, we filter it out.
             warnings.filterwarnings(
                 "ignore",
-                message="Data is not standardized",
+                message=r"Data \(outcome observations\) not standardized",
                 category=InputDataWarning,
             )
             cv_predictions = self._cross_validate(

diff --git a/ax/modelbridge/cross_validation.py b/ax/modelbridge/cross_validation.py
@@ -184,7 +184,7 @@ def cross_validate(
                 # To avoid confusing users with this warning, we filter it out.
                 warnings.filterwarnings(
                     "ignore",
-                    message="Data is not standardized",
+                    message=r"Data \(outcome observations\) not standardized",
                     category=InputDataWarning,
                 )
                 cv_test_predictions = model._cross_validate(

diff --git a/ax/modelbridge/tests/test_base_modelbridge.py b/ax/modelbridge/tests/test_base_modelbridge.py
@@ -225,7 +225,7 @@ def warn_and_return_mock_obs(
             nonlocal called
             called = True
             warnings.warn(
-                "Data is not standardized",
+                "Data (outcome observations) not standardized",
                 InputDataWarning,
                 stacklevel=2,
             )