Update app.py

app.py

@@ -18,9 +18,9 @@ login(token=os.environ["HUGGINGFACEHUB_API_TOKEN"])
 # --- Define Tools ---
 search_tool = DuckDuckGoSearchTool()
 
-# --- Main Async Function ---
+# --- Main Async Function with Progress Logs ---
 async def run_and_submit_all(profile: gr.OAuthProfile | None):
-    # Initialize Agent
+    log_output = ""
     try:
         agent = CodeAgent(
             tools=[search_tool],
@@ -29,23 +29,26 @@ async def run_and_submit_all(profile: gr.OAuthProfile | None):
             verbosity_level=2
         )
     except Exception as e:
-        return f"❌ Agent Initialization Error: {e}", None
+        yield f"❌ Agent Initialization Error: {e}", None, log_output
+        return
 
     space_id = os.getenv("SPACE_ID", "unknown")
     agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
 
-    # Fetch questions
     try:
         response = requests.get(QUESTIONS_URL, timeout=15)
         response.raise_for_status()
         questions = response.json()
         if not questions:
-            return "⚠️ No questions received.", None
+            yield "⚠️ No questions received.", None, log_output
+            return
     except Exception as e:
-        return f"❌ Failed to fetch questions: {e}", None
+        yield f"❌ Failed to fetch questions: {e}", None, log_output
+        return
 
     answers = []
     logs = []
+    loop = asyncio.get_running_loop()
 
     for item in questions:
         task_id = item.get("task_id")
@@ -53,6 +56,9 @@ async def run_and_submit_all(profile: gr.OAuthProfile | None):
         if not task_id or not question:
             continue
 
+        log_output += f"🔍 Solving Task ID: {task_id}...\n"
+        yield None, None, log_output  # Live update
+
         system_prompt = (
             "You are a general AI assistant. I will ask you a question. "
             "Report your thoughts, and finish your answer with the following template: "
@@ -61,7 +67,6 @@ async def run_and_submit_all(profile: gr.OAuthProfile | None):
         full_prompt = system_prompt + f"Question: {question.strip()}"
 
         try:
-            loop = asyncio.get_running_loop()
             result = await loop.run_in_executor(None, lambda: agent(full_prompt))
 
             if isinstance(result, dict) and "final_answer" in result:
@@ -75,16 +80,19 @@ async def run_and_submit_all(profile: gr.OAuthProfile | None):
                 final_answer = str(result).strip()
 
         except Exception as e:
-            print(f"[ERROR] Task {task_id} failed: {e}")
             final_answer = f"AGENT ERROR: {e}"
+            print(f"[ERROR] Task {task_id} failed: {e}")
 
         answers.append({"task_id": task_id, "model_answer": final_answer})
         logs.append({"Task ID": task_id, "Question": question, "Submitted Answer": final_answer})
+        log_output += f"✅ Done: {task_id} — Answer: {final_answer[:60]}\n"
+        yield None, None, log_output  # Live update
 
     valid_answers = [a for a in answers if isinstance(a["task_id"], str) and isinstance(a["model_answer"], str)]
 
     if not valid_answers:
-        return "❌ Agent produced no valid answers.", pd.DataFrame(logs)
+        yield "❌ Agent produced no valid answers.", pd.DataFrame(logs), log_output
+        return
 
     submission = {
         "username": profile.username if profile else "unknown",
@@ -106,10 +114,10 @@ async def run_and_submit_all(profile: gr.OAuthProfile | None):
             f"({result_data.get('correct_count')}/{result_data.get('total_attempted')})\n"
             f"Message: {result_data.get('message', 'No message.')}"
         )
-        return summary, pd.DataFrame(logs)
+        yield summary, pd.DataFrame(logs), log_output
 
     except Exception as e:
-        return f"❌ Submission failed: {e}", pd.DataFrame(logs)
+        yield f"❌ Submission failed: {e}", pd.DataFrame(logs), log_output
 
 # --- Gradio UI ---
 with gr.Blocks() as demo:
@@ -117,15 +125,16 @@ with gr.Blocks() as demo:
     gr.Markdown("""
     - Log in with your Hugging Face account.
     - Click the button below to run the agent and submit the answers.
-    - Wait for the final score to appear.
+    - Watch the log to see which question is being solved in real-time.
     """)
 
     gr.LoginButton()
     run_button = gr.Button("🚀 Run Evaluation & Submit")
-    status = gr.Textbox(label="Status", lines=6)
+    status = gr.Textbox(label="Final Status", lines=6)
     table = gr.DataFrame(label="Answer Log")
+    progress_log = gr.Textbox(label="Live Progress Log", lines=10, interactive=False)
 
-    run_button.click(fn=run_and_submit_all, outputs=[status, table])
+    run_button.click(fn=run_and_submit_all, outputs=[status, table, progress_log])
 
 # --- Launch ---
 if __name__ == "__main__":