update

question_recommendation/run_labeling.py

@@ -6,9 +6,10 @@ import pandas as pd
 import json
 from tqdm import tqdm
 
+
 def chat(question, answer):
-    client = OpenAI(api_key=os.getenv('OPENAI_API_KEY'))
-    
+    client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
+
     completion = client.chat.completions.create(
         model="gpt-4o",
         messages=[
@@ -16,45 +17,51 @@ def chat(question, answer):
                 "role": "system",
                 "content": (
                     "당신은 대화를 오래 이어가기위해 사용자에게 질문을 추천해주는 봇입니다."
-                    "즉, 사용자가 인공지능 어시스턴트에게 추가 질문을 진행합니다."
+                    "사용자는 인공지능 어시스턴트에게 추가 질문을 진행합니다."
                     "현재 대화에서 사용자가 어시스턴트에게 질문하면 좋을 것 같은 3가지 질의를 추천해주세요."
                     "출력 형식은 '1.str\n2.str\n3.str' 입니다."
                     "출력 예시는 다음과 같습니다."
                     "1. 메뉴에 있는 음료 중 가장 비싼건 무엇인가요?\n2. 메뉴에서 따뜻한 음료와 차가운 음료 옵션을 모두 제공하는 항목은 무엇인가요? \n3. 스페셜티 커피는 어떤 종류가 있나요?"
-                )
+                ),
             },
             {
                 "role": "user",
-                "content": (f"사용자:{question}\n 어시스턴트:{answer}"
-                "위 내용에 대해 어시스턴트에게 추가로 물어볼만한 질의를 3개 추천해주세요."),
-            }
+                "content": (
+                    f"사용자:{question}\n 어시스턴트:{answer}"
+                    "위 내용에 대해 어시스턴트에게 추가로 물어볼만한 질의를 3개 추천해주세요."
+                ),
+            },
         ],
     )
     return completion.choices[0].message.content.strip()
 
+
 def add_recommendation(input_file, output_file):
-    with open(input_file, 'r', encoding='utf-8') as f:
+    with open(input_file, "r", encoding="utf-8") as f:
         data = json.load(f)
-    
+
     total_entries = len(data)
-    
+
     for i, entry in enumerate(tqdm(data)):
         question = entry.get("질문", "")
         answer = entry.get("답변", "")
         recommendation = chat(question, answer)
         entry["추천질의"] = recommendation
-        tqdm.write(f"{i + 1}/{total_entries} 완료. ({(i + 1) / total_entries * 100:.2f}%)")
-    
-    with open(output_file, 'w', encoding='utf-8') as f:
+        tqdm.write(
+            f"{i + 1}/{total_entries} 완료. ({(i + 1) / total_entries * 100:.2f}%)"
+        )
+
+    with open(output_file, "w", encoding="utf-8") as f:
         json.dump(data, f, ensure_ascii=False, indent=4)
 
+
 def concat_jsons(before_file_path, additional_file_path, after_file_path):
-    with open(before_file_path, 'r', encoding='utf-8') as file_01:
+    with open(before_file_path, "r", encoding="utf-8") as file_01:
         before_json = json.load(file_01)
-        
-    with open(additional_file_path, 'r', encoding='utf-8') as file_02:
+
+    with open(additional_file_path, "r", encoding="utf-8") as file_02:
         additional_json = json.load(file_02)
-    
+
     for item in tqdm(additional_json):
         new_entry = {
             "질문": item.get("질문", ""),
@@ -65,34 +72,33 @@ def concat_jsons(before_file_path, additional_file_path, after_file_path):
             "세션ID": None,
             "모델타입": None,
             "사용자 평가": None,
-            "모인ID": None
+            "모인ID": None,
         }
         before_json.append(new_entry)
 
-    with open(after_file_path, 'w', encoding='utf-8') as file:
+    with open(after_file_path, "w", encoding="utf-8") as file:
         json.dump(before_json, file, ensure_ascii=False, indent=4)
 
 
-
-
 def testing():
     question = "안녕"
     answer = "안녕하세요. 무엇을 도와드릴까요?"
-    
+
     recommendation = chat(question, answer)
     print(f"\033[94m{recommendation}\033[0m")
-    
+
     return None
 
 
 if __name__ == "__main__":
     testing()
-    
-    add_recommendation(input_file="QR_v1.4a.json",
-                       output_file="QR_v1_4b.json")
 
-    concat_jsons(before_file_path = "QR_v1.4.json", 
-                 additional_file_path = "QR_v1.4b.json", 
-                 after_file_path= "QR_v1.5.json")
+    add_recommendation(input_file="QR_v1.4a.json", output_file="QR_v1_4b.json")
+
+    concat_jsons(
+        before_file_path="QR_v1.4.json",
+        additional_file_path="QR_v1.4b.json",
+        after_file_path="QR_v1.5.json",
+    )
 
     print(f"\033[94m ##DONE## \033[0m")