Add PED-ANOVA as the first option for importance evaluator

optuna_dashboard/_importance.py

@@ -26,6 +26,16 @@
     FastFanovaImportanceEvaluator = None  # type: ignore
 
 
+try:
+    from optuna.importance import PedAnovaImportanceEvaluator
+except ImportError:
+    _logger.warning(f"optuna>=3.6.0 is required for PedAnovaImportanceEvaluator.")
+    PedAnovaImportanceEvaluator = None  # type: ignore
+except Exception as e:
+    _logger.warning(f"Skipping to use PedAnovaImportanceEvaluator due to {e}.")
+    PedAnovaImportanceEvaluator = None  # type: ignore
+
+
 if TYPE_CHECKING:
     from typing import Callable
     from typing import Optional
@@ -64,6 +74,12 @@
     *,
     target: Optional[Callable[[FrozenTrial], float]] = None,
 ) -> dict[str, float]:
+    if PedAnovaImportanceEvaluator is not None:
+        # TODO(nabenabe0928): We might want to pass baseline_quantile as an argument in the future.
+        return get_param_importances(
+            study, target=target, evaluator=PedAnovaImportanceEvaluator()
+        )
+
     if FastFanovaImportanceEvaluator is not None:
         try:
             evaluator = FastFanovaImportanceEvaluator(completed_trials=completed_trials)