degbu

generic_ner.py

@@ -274,15 +274,7 @@ class MultitaskTokenClassificationPipeline(Pipeline):
         """Check if entity1 is fully within the bounds of entity2."""
         return entity1["start"] >= entity2["start"] and entity1["end"] <= entity2["end"]
 
-    def postprocess_entities(self, all_entities, text_sentence):
-        # Collect all entities in one list for processing
-
-        # for entity in ner_results[key]:
-        #     print(
-        #         text_sentence[entity["start"] : entity["end"]],
-        #         "--",
-        #         entity["word"],
-        #     )
+    def postprocess_entities(self, all_entities):
 
         # Sort entities by start position, then by end position (to handle nested structures)
         all_entities.sort(key=lambda x: (x["start"], -x["end"]))
@@ -363,12 +355,13 @@ class MultitaskTokenClassificationPipeline(Pipeline):
             if key not in ["NE-COARSE-LIT"]:
                 all_entities.extend(entities[key])
 
-        postprocessed_entities = self.postprocess_entities(all_entities, text_sentence)
-        print("After 1:")
-        pprint(postprocessed_entities)
+        print("Skipping 1")
+        # all_entities = self.postprocess_entities(all_entities, text_sentence)
+        # print("After 1:")
+        # pprint(all_entities)
         # Attach "comp.function" entities to the closest non-"comp.function" entity
-        postprocessed_entities = attach_comp_to_closest(postprocessed_entities)
+        all_entities = attach_comp_to_closest(all_entities)
         print("After 2:")
-        pprint(postprocessed_entities)
+        pprint(all_entities)
         print("\n")
-        return postprocessed_entities
+        return all_entities