Adding KLD wrapper.

src/qp/metrics/base_metric_classes.py

@@ -28,26 +28,26 @@ class MetricInputType(enum.Enum):
     point_to_dist = 4
 
 
-    def uses_distribution_for_estimate(self):
+    def uses_distribution_for_estimate(self) -> bool:
         return self in [
             MetricInputType.single_ensemble,
             MetricInputType.dist_to_point,
             MetricInputType.dist_to_dist,
         ]
 
-    def uses_distribution_for_reference(self):
+    def uses_distribution_for_reference(self) -> bool:
         return self in [
             MetricInputType.dist_to_dist,
             MetricInputType.point_to_dist,
         ]
 
-    def uses_point_for_estimate(self):
+    def uses_point_for_estimate(self) -> bool:
         return self in [
             MetricInputType.point_to_dist,
             MetricInputType.point_to_dist,
         ]
 
-    def uses_point_for_reference(self):
+    def uses_point_for_reference(self) -> bool:
         return self in [
             MetricInputType.dist_to_point,
             MetricInputType.point_to_point,
@@ -80,29 +80,20 @@ def __init__(self, limit:tuple=(0.0, 3.0), dx:float=0.01) -> None:
         self._limit = limit
         self._dx = dx
 
-    def initialize(self) -> None:
-        pass
-
-    def evaluate(self) -> None:
-        pass
-
-    def finalize(self) -> None:
-        pass
-
     @classmethod
-    def uses_distribution_for_estimate(cls):
+    def uses_distribution_for_estimate(cls) -> bool:
         return cls.metric_input_type.uses_distribution_for_estimate()
 
     @classmethod
-    def uses_distribution_for_reference(cls):
+    def uses_distribution_for_reference(cls) -> bool:
         return cls.metric_input_type.uses_distribution_for_reference()
 
     @classmethod
-    def uses_point_for_estimate(cls):
+    def uses_point_for_estimate(cls) -> bool:
         return cls.metric_input_type.uses_point_for_estimate()
 
     @classmethod
-    def uses_point_for_reference(cls):
+    def uses_point_for_reference(cls) -> bool:
         return cls.metric_input_type.uses_point_for_reference()
 
 
@@ -121,6 +112,15 @@ class DistToPointMetric(BaseMetric):
 
     metric_input_type = MetricInputType.dist_to_point
 
+    def initialize(self, **kwargs):
+        raise NotImplementedError()
+
+    def evaluate(self, estimate, reference, **kwargs):
+        raise NotImplementedError()
+
+    def finalize(self, **kwargs):
+        raise NotImplementedError()
+
 
 class PointToPointMetric(BaseMetric):
 

src/qp/metrics/concrete_metric_classes.py

@@ -0,0 +1,23 @@
+from qp.metrics.base_metric_classes import MetricOuputType, DistToDistMetric
+from qp.metrics.metrics import calculate_kld
+
+class KLDMetric(DistToDistMetric):
+    """Class wrapper around the KLD metric
+    """
+
+    metric_name = "kld"
+    metric_output_type = MetricOuputType.one_value_per_distribution
+
+    def __init__(self, limits:tuple=(0.0, 3.0), dx:float=0.01) -> None:
+        super().__init__()
+        self._limits = limits
+        self._dx = dx
+
+    def initialize(self) -> None:
+        pass
+
+    def evaluate(self, estimate, reference) -> None:
+        return calculate_kld(estimate, reference, self._limits, self._dx)
+
+    def finalize(self) -> None:
+        pass