Update app.py

app.py

@@ -463,17 +463,43 @@ def bot(history, choice, tts_choice, retrieval_mode, model_choice):
 
 import re
 
+# def clean_response(response_text):
+#     # Remove system and user tags
+#     response_text = re.sub(r'<\|system\|>.*?<\|end\|>', '', response_text, flags=re.DOTALL)
+#     response_text = re.sub(r'<\|user\|>.*?<\|end\|>', '', response_text, flags=re.DOTALL)
+#     response_text = re.sub(r'<\|assistant\|>', '', response_text, flags=re.DOTALL)
+#     #-------
+#     response_text = re.sub(r'\\u[0-9A-Fa-f]{4}', '', response_text)
+#     # Add spaces between words and dates (e.g., "born04/04/1963" becomes "born 04/04/1963")
+#     response_text = re.sub(r'([a-zA-Z])(\d)', r'\1 \2', response_text)
+#     response_text = re.sub(r'(\d)([a-zA-Z])', r'\1 \2', response_text)
+#     #--------
+#     # Clean up the text by removing extra whitespace
+#     cleaned_response = response_text.strip()
+#     cleaned_response = re.sub(r'\s+', ' ', cleaned_response)
+
+#     # Ensure the response is conversational and organized
+#     cleaned_response = cleaned_response.replace('1.', '\n1.').replace('2.', '\n2.').replace('3.', '\n3.').replace('4.', '\n4.').replace('5.', '\n5.')
+
+#     return cleaned_response
+
+
 def clean_response(response_text):
     # Remove system and user tags
     response_text = re.sub(r'<\|system\|>.*?<\|end\|>', '', response_text, flags=re.DOTALL)
     response_text = re.sub(r'<\|user\|>.*?<\|end\|>', '', response_text, flags=re.DOTALL)
     response_text = re.sub(r'<\|assistant\|>', '', response_text, flags=re.DOTALL)
-    #-------
+    
+    # Remove 'Document(metadata=...' part
+    response_text = re.sub(r'Document\(metadata=\{.*?\},page_content=', '', response_text, flags=re.DOTALL)
+    
+    # Remove any unwanted escape characters like \u and \u00
     response_text = re.sub(r'\\u[0-9A-Fa-f]{4}', '', response_text)
+
     # Add spaces between words and dates (e.g., "born04/04/1963" becomes "born 04/04/1963")
     response_text = re.sub(r'([a-zA-Z])(\d)', r'\1 \2', response_text)
     response_text = re.sub(r'(\d)([a-zA-Z])', r'\1 \2', response_text)
-    #--------
+
     # Clean up the text by removing extra whitespace
     cleaned_response = response_text.strip()
     cleaned_response = re.sub(r'\s+', ' ', cleaned_response)
@@ -485,6 +511,7 @@ def clean_response(response_text):
 
 
 
+
 # Define a new template specifically for GPT-4o-mini in VDB Details mode
 gpt4o_mini_template_details = f"""
 As a highly specialized assistant, I provide precise, detailed, and informative responses. On this bright day of {current_date}, I'm equipped to assist with all your queries about Birmingham, Alabama, offering detailed insights tailored to your needs.