updated to not use labels

app.py

@@ -34,46 +34,75 @@ def extract_features(tokens, labels):
     partial_date_pattern = r"^\d{1,2}/?$|^[/-]$"  # Matches partial date components like "12", "/", "02/", etc.
 
 
-    
-    # Loop through tokens and labels 
-    for token, label in zip(tokens, labels): 
-        print(f"Debug -- Potentially creating date,, token: {token} label: {label}")
-        
-        if label == 'LABEL_1':
-            # Check for partial date fragments (like '12' or '/')
-            if re.match(date_pattern, current_date):
-                merged_entities.append((current_date, 'date'))
-                print(f"Debug -- Complete date added: {token}")
-                current_date = ""  # Reset for next entity
-            # If the accumulated entity matches a full date
-            elif re.match(partial_date_pattern, token):
-                print(f"Debug -- Potentially building date: Token Start {token} After Token")
-                current_date += token # Append token to the current entity
-            else: 
-                # No partial or completed patterns are detected, but it's still LABEL_1
-                # If there were any accumulated data so far
-                if current_date: 
-                    merged_entities.append((current_date, 'date'))
-                    print(f"Debug -- Date finalized: {current_date}")
-                    current_date = "" # Reset
-                
-                merged_entities.append((token, label))
+    # This is a label AGNOSTIC approach 
+    for token, label in zip(tokens, labels):  
+        print(f"Debug -- Processing token: {token}")
+
+        # Check complete date pattern 
+        if re.match(date_pattern, current_date): 
+            merged_entities.append((current_date, 'date'))
+            print(f"Debug -- Completed date added: {current_date}")
+            current_date = ""
+        # Check if the token matches a partial date pattern
+        elif re.match(partial_date_pattern, token):
+            print(f"Debug -- Potentially building date: Token Start {token} After Token")
+            current_date += LayoutLMv3ForTokenClassification
         else: 
-            # These are LABEL_0, supposedly trash but keep them for now
-            if current_date:  # If there's a leftover date fragment, add it first
+            if current_date: 
                 merged_entities.append((current_date, 'date'))
-                print(f"Debug -- Finalizing leftover date added: {current_date}")
-                current_date = ""  # Reset
+                print(f"Debug -- Date finalized: {current_date}")
+                current_date = ""
 
-            # Append LABEL_0 token
-            print(f"Debug -- Appending LABEL_0 Token: Token Start {token} Token After")
-            merged_entities.append((token, label))
+            print(f"Debug -- Appending non-date Token: {token}")
+            merged_entities.append((token, 'other'))
 
-    if current_date:
+    if current_date: 
         print(f"Debug -- Dangling leftover date added: {current_date}")
         merged_entities.append((current_date, 'date'))
 
     return merged_entities
+
+    # NOTE: labels aren't being applied properly ... This is the LABEL approach 
+    #
+    # Loop through tokens and labels 
+    #for token, label in zip(tokens, labels): 
+    #   print(f"Debug -- Potentially creating date,, token: {token} label: {label}")
+    #    
+    #    if label == 'LABEL_1':
+    #        # Check for partial date fragments (like '12' or '/')
+    #        if re.match(date_pattern, current_date):
+    #            merged_entities.append((current_date, 'date'))
+    #            print(f"Debug -- Complete date added: {token}")
+    #            current_date = ""  # Reset for next entity
+    #        # If the accumulated entity matches a full date
+    #        elif re.match(partial_date_pattern, token):
+    #            print(f"Debug -- Potentially building date: Token Start {token} After Token")
+    #            current_date += token # Append token to the current entity
+    #        else: 
+    #            # No partial or completed patterns are detected, but it's still LABEL_1
+    #            # If there were any accumulated data so far
+    #            if current_date: 
+    #                merged_entities.append((current_date, 'date'))
+    #                print(f"Debug -- Date finalized: {current_date}")
+    #                current_date = "" # Reset
+    #            
+    #            merged_entities.append((token, label))
+    #    else: 
+    #        # These are LABEL_0, supposedly trash but keep them for now
+    #        if current_date:  # If there's a leftover date fragment, add it first
+    #            merged_entities.append((current_date, 'date'))
+    #            print(f"Debug -- Finalizing leftover date added: {current_date}")
+    #            current_date = ""  # Reset
+#
+#            # Append LABEL_0 token
+#            print(f"Debug -- Appending LABEL_0 Token: Token Start {token} Token After")
+#            merged_entities.append((token, label))
+#
+#    if current_date:
+#        print(f"Debug -- Dangling leftover date added: {current_date}")
+#        merged_entities.append((current_date, 'date'))
+#
+#    return merged_entities
     
     
 # process the image in the correct format