Update app.py

app.py

@@ -8,6 +8,8 @@ from datasets import load_dataset
 import pandas as pd
 from fuzzywuzzy import process
 
+
+
 # 현재 작업 디렉토리 출력
 print("Current Working Directory:", os.getcwd())
 
@@ -28,16 +30,12 @@ else:
 def load_optimized_dataset(data_files):
     data_frames = [pd.read_csv(file) for file in data_files]
     full_data = pd.concat(data_frames, ignore_index=True)
+    # 확인을 위한 데이터 샘플 출력
+    logging.debug(f"Data sample: {full_data[['사건명', '사건번호', '판시사항']].head()}")
     # 사건명을 키로 하고 사건번호와 전문을 저장하는 딕셔너리 생성
     name_to_number = full_data.groupby('사건명')['사건번호'].apply(list).to_dict()
     summary_to_number = full_data.groupby('판시사항')['사건번호'].apply(list).to_dict()
     number_to_fulltext = full_data.set_index('사건번호')['전문'].to_dict()
-
-    # 디버깅용 로깅
-    logging.debug(f"Sample entries in name_to_number: {list(name_to_number.items())[:3]}")
-    logging.debug(f"Sample entries in summary_to_number: {list(summary_to_number.items())[:3]}")
-    logging.debug(f"Sample entries in number_to_fulltext: {list(number_to_fulltext.items())[:3]}")
-
     return name_to_number, summary_to_number, number_to_fulltext
 
 name_to_number, summary_to_number, number_to_fulltext = load_optimized_dataset(data_files)
@@ -112,7 +110,7 @@ async def generate_response(message):
     user_mention = message.author.mention
 
     # 유사한 사건명 및 판시사항 찾기
-    matched_case_names = process.extractBests(user_input, all_case_names, limit=3, score_cutoff=70)  # score_cutoff 값 조정
+    matched_case_names = process.extractBests(user_input, all_case_names, limit=3, score_cutoff=70)
     matched_case_summaries = process.extractBests(user_input, all_case_summaries, limit=3, score_cutoff=70)
 
     logging.debug(f"Matched case names: {matched_case_names}")
@@ -121,15 +119,15 @@ async def generate_response(message):
     if matched_case_names:
         case_numbers = []
         for case_name, score in matched_case_names:
-            case_numbers.extend(name_to_number[case_name])
-        case_numbers = list(set(case_numbers))  # 중복 제거
+            case_numbers.extend(name_to_number.get(case_name, []))
+        case_numbers = list(set(case_numbers))
         case_numbers_str = "\n".join(case_numbers)
         system_message = f"{user_mention}, '{user_input}'와 유사한 사건명의 사건번호는 다음과 같습니다:\n{case_numbers_str}"
     elif matched_case_summaries:
         case_numbers = []
         for case_summary, score in matched_case_summaries:
-            case_numbers.extend(summary_to_number[case_summary])
-        case_numbers = list(set(case_numbers))  # 중복 제거
+            case_numbers.extend(summary_to_number.get(case_summary, []))
+        case_numbers = list(set(case_numbers))
         case_numbers_str = "\n".join(case_numbers)
         system_message = f"{user_mention}, '{user_input}'와 유사한 판시사항의 사건번호는 다음과 같습니다:\n{case_numbers_str}"
     elif user_input in number_to_fulltext: