Consistent metric tracker

allennlp/training/metric_tracker.py

@@ -31,13 +31,13 @@ def __init__(
         metric_name: Union[str, List[str]],
         patience: Optional[int] = None,
     ) -> None:
-        self._best_so_far: Optional[float] = None
         self._patience = patience
+        self._best_so_far: Optional[float] = None
         self._epochs_with_no_improvement = 0
         self._is_best_so_far = True
-        self.best_epoch_metrics: Dict[str, float] = {}
         self._epoch_number = 0
         self.best_epoch: Optional[int] = None
+        self.best_epoch_metrics: Dict[str, float] = {}
 
         if isinstance(metric_name, str):
             metric_name = [metric_name]
@@ -59,32 +59,34 @@ def clear(self) -> None:
         self._is_best_so_far = True
         self._epoch_number = 0
         self.best_epoch = None
+        self.best_epoch_metrics.clear()
 
     def state_dict(self) -> Dict[str, Any]:
         """
         A `Trainer` can use this to serialize the state of the metric tracker.
         """
         return {
             "best_so_far": self._best_so_far,
-            "patience": self._patience,
             "epochs_with_no_improvement": self._epochs_with_no_improvement,
             "is_best_so_far": self._is_best_so_far,
-            "best_epoch_metrics": self.best_epoch_metrics,
             "epoch_number": self._epoch_number,
             "best_epoch": self.best_epoch,
+            "best_epoch_metrics": self.best_epoch_metrics,
         }
 
     def load_state_dict(self, state_dict: Dict[str, Any]) -> None:
         """
         A `Trainer` can use this to hydrate a metric tracker from a serialized state.
         """
         self._best_so_far = state_dict["best_so_far"]
-        self._patience = state_dict["patience"]
         self._epochs_with_no_improvement = state_dict["epochs_with_no_improvement"]
         self._is_best_so_far = state_dict["is_best_so_far"]
-        self.best_epoch_metrics = state_dict["best_epoch_metrics"]
         self._epoch_number = state_dict["epoch_number"]
         self.best_epoch = state_dict["best_epoch"]
+
+        # Even though we don't promise backwards compatibility for the --recover flag,
+        # it's particularly easy and harmless to provide it here, so we do it.
+        self.best_epoch_metrics = state_dict.get("best_epoch_metrics", {})
 
     def add_metrics(self, metrics: Dict[str, float]) -> None:
         """
@@ -103,13 +105,13 @@ def add_metrics(self, metrics: Dict[str, float]) -> None:
         new_best = (self._best_so_far is None) or (combined_score > self._best_so_far)
 
         if new_best:
-            self.best_epoch = self._epoch_number
-            self._is_best_so_far = True
             self._best_so_far = combined_score
             self._epochs_with_no_improvement = 0
+            self._is_best_so_far = True
+            self.best_epoch = self._epoch_number
         else:
-            self._is_best_so_far = False
             self._epochs_with_no_improvement += 1
+            self._is_best_so_far = False
         self._epoch_number += 1
 
     def is_best_so_far(self) -> bool: