refactor to reinit class

user-friendly-metrics.py

@@ -60,6 +60,20 @@ Args:
 class UserFriendlyMetrics(evaluate.Metric):
     """TODO: Short description of my evaluation module."""
 
+    def __init__(self, 
+        iou_threshold = 1e-10,
+        filter = {"name": "area", "ranges": [("all", [0, 1e5**2])]},
+        payload: Payload = None,,
+        recognition_thresholds: list = [0.3, 0.5, 0.8],
+        **kwargs
+        ):
+        super().__init__(**kwargs)
+
+        self.iou_threshold = iou_threshold
+        self.filter = filter
+        self.payload = payload
+        self.recognition_thresholds = recognition_thresholds
+
     def _info(self):
         # TODO: Specifies the evaluate.EvaluationModuleInfo object
         return evaluate.MetricInfo(
@@ -89,23 +103,59 @@ class UserFriendlyMetrics(evaluate.Metric):
         # TODO: Download external resources if needed
         pass
 
-    def compute_from_payload(self, payload, **kwargs):
-        return self._compute(payload, **kwargs)
+    # def compute_from_payload(self, payload, **kwargs):
+        
 
-    def _compute(
-        self,
-        payload,
-        max_iou: float = 0.5,
-        filter={},
-        recognition_thresholds=[0.3, 0.5, 0.8],
-        debug: bool = False,
-    ):
-        """Returns the scores"""
-        # TODO: Compute the different scores of the module
+    def _compute(self, **kwargs):
+        
         return calculate_from_payload(
-            payload, max_iou, filter, recognition_thresholds, debug
+            self.payload,
+            self.iou_threshold,
+            self.filter,
+            self.recognition_thresholds,
+            **kwargs
         )
-        # return calculate(predictions, references, max_iou)
+
+    def compute_from_payload(self, 
+        payload: Payload, 
+        **kwargs):
+
+        results = {}
+
+        for model_name in payload.models:
+            results[model_name] = {"overall": {}, "per_sequence": {}}
+            
+            # per-sequence loop
+            for seq_name, sequence in payload.sequences.items():
+                # create new payload only with specific sequence and model
+                sequence_payload = Payload(
+                    dataset=payload.dataset,
+                    gt_field_name=payload.gt_field_name,
+                    models=[model_name],
+                    sequences={seq_name: sequence}
+                )
+                module = UserFriendlyMetrics(
+                    iou_threshold=self.iou_threshold,
+                    filter=self.filter,
+                    payload=sequence_payload
+                    recognition_thresholds=self.recognition_thresholds
+                )
+                results[model_name]["per_sequence"][seq_name] = module.compute()[model_name]["metrics"]
+
+            # overall per-model loop
+            model_payload = Payload(
+                    dataset=payload.dataset,
+                    gt_field_name=payload.gt_field_name,
+                    models=[model_name],
+                    sequences=payload.sequences
+                )
+            module = UserFriendlyMetrics(
+                iou_threshold=self.iou_threshold,
+                filter=self.filter,
+                payload=model_payload
+                recognition_thresholds=self.recognition_thresholds
+            )
+            results[model_name]["overall"] = module.compute()[model_name]["metrics"]
 
     def wandb(
         self,