Update app.py

app.py

@@ -10,29 +10,29 @@ from transformers import pipeline
 
 pipe = pipeline("text-generation", model="jslin09/gemma2-2b-ner", device="cuda")
 
-elements = {'LEO_SOC': ('犯罪主體', 'Subject of Crime'),
-            'LEO_VIC': ('客體', 'Victim'),
-            'LEO_ACT': ('不法行為', 'Behavior'),
-            'LEO_SLE': ('主觀要件', 'Subjective Legal Element of the Offense'),
-            'LEO_CAU': ('因果關係', 'Causation'),
-            'LEO_ROH': ('危害結果', 'Result of Hazard'),
-            'LEO_ATP': ('未遂', 'Attempted'),
-            'LEO_ACC': ('既遂', 'Accomplished'),
-            'LEO_ABA': ('中止', 'Abandonment'),
-            'LEO_PRP': ('預備', 'Preparation'),
-            'ILG_GLJ': ('阻卻違法事由', 'Ground of Legal Justification'),
-            'ILG_SDE': ('正當防衛', 'Self-Defense'),
-            'ILG_NEC': ('緊急避難', 'Emergency Avoidance'),
-            'CUL_INS': ('心神喪失', 'Insane'),
-            'CUL_FBD': ('精神耗弱', 'Feebleminded'),
-            'CUL_ALC': ('原因自由行為', 'Actio Libera in Causa'),
-            'CUL_RPS': ('責任能力', 'Responsibility'),
-            'CUL_ANP': ('期待可能性', 'Anticipated Possibility'),
-            'CUL_GUM': ('犯罪意識', 'Guilty Mind')
-           }
+all_elements = {'LEO_SOC': ('犯罪主體', 'Subject of Crime'),
+                'LEO_VIC': ('客體', 'Victim'),
+                'LEO_ACT': ('不法行為', 'Behavior'),
+                'LEO_SLE': ('主觀要件', 'Subjective Legal Element of the Offense'),
+                'LEO_CAU': ('因果關係', 'Causation'),
+                'LEO_ROH': ('危害結果', 'Result of Hazard'),
+                'LEO_ATP': ('未遂', 'Attempted'),
+                'LEO_ACC': ('既遂', 'Accomplished'),
+                'LEO_ABA': ('中止', 'Abandonment'),
+                'LEO_PRP': ('預備', 'Preparation'),
+                'ILG_GLJ': ('阻卻違法事由', 'Ground of Legal Justification'),
+                'ILG_SDE': ('正當防衛', 'Self-Defense'),
+                'ILG_NEC': ('緊急避難', 'Emergency Avoidance'),
+                'CUL_INS': ('心神喪失', 'Insane'),
+                'CUL_FBD': ('精神耗弱', 'Feebleminded'),
+                'CUL_ALC': ('原因自由行為', 'Actio Libera in Causa'),
+                'CUL_RPS': ('責任能力', 'Responsibility'),
+                'CUL_ANP': ('期待可能性', 'Anticipated Possibility'),
+                'CUL_GUM': ('犯罪意識', 'Guilty Mind')
+               }
 
-secret = os.environ.get("HF_TOKEN")
-llm_server = os.environ.get("REMOTE_LLM_SERVER")
+#secret = os.environ.get("HF_TOKEN")
+#llm_server = os.environ.get("REMOTE_LLM_SERVER")
 legal_element_tags = ['LEO_SOC', 'LEO_VIC', 'LEO_ACT', 'LEO_SLE', 'LEO_CAU', 'LEO_ROH']
 
 def get_prompt(content, tag, tag_name):
@@ -135,9 +135,6 @@ def ner_extract(text, le_list):
             ner_list.append(ner_dict)
             index = index + 1
     return ner_list
-@spaces.GPU
-def dummy():
-    return None
 
 @spaces.GPU(duration=180)
 def le_ner(content, legal_element_tags=legal_element_tags):