track

app.py

@@ -76,10 +76,15 @@ def get_prompt_plain(context: str, question: str) -> str:
     return f"""
 <<START>>
 You are a responsible person for answering Inha University (South Korea) information. Using the context below, answer within 300 tokens.
-Create interactive, well-structured answers using bullet points, bold text, and proper formatting to make the information, Provide concise, answer-oriented, clear and easy to read.
+Create interactive, well-structured answers using bullet points, bold text, and proper formatting to make the information concise, answer-oriented, clear and easy to read.
 Do not repeat the prompt text in your output.
 And when context doesn't provide what user hasn't asked, don't mention it. Instead, just say in polite way you don't know it 
 And in context text, there always will be link where this info is taken. at the end of your response, say that user can visit this link for official information. Don't forget to mention link
+and at the end of response don't forget to say polite words like "have a nice day" "Have a wonderful day ""Have an awesome day" Stay awesome, or Make it a great day
+Go make some magic happen
+Here's to a fantastic day ahead
+May your day be filled with good things
+Hope something wonderful happens to you today . in general, based on question, adjust them
 Context:
 "{context}"
 
@@ -106,27 +111,100 @@ def rag_answer(question: str, collection) -> str:
     context = retrieve_context(question, collection, top_k=2)
     return generate_agent_answer(context, question)
 
-# gradio interface code below
+from datasets import Dataset, load_dataset
+from huggingface_hub import HfApi
+from datetime import datetime
+import pandas as pd
+import uuid
+import os
+
+
+HF_TOKEN = os.getenv("HF_TOKEN")  # Set this in Space settings
+DATASET_NAME = "Jasur05/inha-chat-logs" 
+
+def log_to_dataset(question, answer, response_time_ms=None):
+    """Log interaction to HuggingFace Dataset (permanent storage)"""
+    try:
+        
+        new_data = {
+            "timestamp": [datetime.now().isoformat()],
+            "session_id": [str(uuid.uuid4())[:8]],
+            "question": [question],
+            "answer": [answer],
+            "response_time_ms": [response_time_ms or 0]
+        }
+        
+      
+        try:
+            existing_dataset = load_dataset(DATASET_NAME, split="train", token=HF_TOKEN)
+            existing_df = existing_dataset.to_pandas()
+            
+       
+            new_df = pd.DataFrame(new_data)
+            combined_df = pd.concat([existing_df, new_df], ignore_index=True)
+            
+        except Exception:
+     
+            combined_df = pd.DataFrame(new_data)
+        
+
+        dataset = Dataset.from_pandas(combined_df)
+        dataset.push_to_hub(
+            DATASET_NAME,
+            token=HF_TOKEN,
+            private=True  # Keep your logs private
+        )
+        
+        print(f"✅ Logged to dataset: {question[:50]}...")
+        
+    except Exception as e:
+        print(f"❌ Dataset logging failed: {e}")
+        # Fallback to local file
+        log_to_local_file(question, answer, response_time_ms)
+
+def log_to_local_file(question, answer, response_time_ms):
+  
+    try:
+        data = {
+            "timestamp": datetime.now().isoformat(),
+            "question": question,
+            "answer": answer,
+            "response_time_ms": response_time_ms
+        }
+        
+
+        df = pd.DataFrame([data])
+        file_exists = os.path.exists("backup_logs.csv")
+        df.to_csv("backup_logs.csv", mode='a', header=not file_exists, index=False)
+        
+    except Exception as e:
+        print(f"❌ Backup logging failed: {e}")
+
 def answer_question(question):
-    """
-    Main function that processes the question and returns the answer
-    """
+   
+    start_time = datetime.now()
+    
     if not question.strip():
-        return "Please enter a question about Inha University."
+        answer = "Please enter a question about Inha University."
+        response_time_ms = 0
+    else:
+        try:
+            answer = rag_answer(question, collection)
+            response_time_ms = (datetime.now() - start_time).total_seconds() * 1000
+        except Exception as e:
+            answer = f"Sorry, I encountered an error: {str(e)}"
+            response_time_ms = (datetime.now() - start_time).total_seconds() * 1000
+
+    log_to_dataset(question, answer, response_time_ms)
     
-    try:
-        answer = rag_answer(question, collection)
-        return answer
-    except Exception as e:
-        return f"Sorry, I encountered an error: {str(e)}"
+    return answer
+
 
-# ─── 6. Gradio Frontend ─────────────────────────────────────────────────────
-# Create the Gradio interface
 demo = gr.Interface(
     fn=answer_question,
     inputs=gr.Textbox(
         label="Ask me anything about Inha University SGCS…",
-        placeholder="e.g. How many Major Required credits should I take for graduation? ",
+        placeholder="e.g. How many Major Required credits should I take for graduation?",
         lines=2
     ),
     outputs=gr.Markdown(
@@ -134,7 +212,7 @@ demo = gr.Interface(
         show_copy_button=True
     ),
     title="📚 Inha University SGCS Info Assistant",
-    description="Get answers to your questions about Inha University SGCS .",
+    description="Get answers to your questions about Inha University SGCS.",
     theme=gr.themes.Soft(),
     examples=[
         ["What classes should I normally take as 3nd semester ISE student?"],
@@ -143,7 +221,8 @@ demo = gr.Interface(
     ]
 )
 
-
+if __name__ == "__main__":
+    demo.launch()
 
 if __name__ == "__main__":
     demo.launch(