Update app.py

app.py

@@ -1,7 +1,129 @@
+import os
 import gradio as gr
+import requests
+import pandas as pd
+from smolagents import CodeAgent, HfApiModel, DuckDuckGoSearchTool
 
-def greet(name):
-    return "Hello " + name + "!!"
+# --- Constants ---
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
-demo.launch()
+# --- Advanced Agent Definition ---
+class BasicAgent:
+    def __init__(self):
+        # Using a powerful model that can handle logic and tool use
+        self.model = HfApiModel(model_id="Qwen/Qwen2.5-72B-Instruct")
+        
+        # We give the agent search capabilities and a Python interpreter
+        self.agent = CodeAgent(
+            tools=[DuckDuckGoSearchTool()], 
+            model=self.model,
+            add_base_tools=True 
+        )
+        print("Advanced smolagent initialized.")
+
+    def __call__(self, question: str) -> str:
+        # Prompting for Exact Match scoring
+        clean_prompt = (
+            f"Question: {question}\n\n"
+            "Instructions: Solve the question above. Provide ONLY the final answer "
+            "value without any explanation, units, or extra text. "
+            "Do not include 'FINAL ANSWER' in your response."
+        )
+        
+        try:
+            # The agent will reason and use tools if necessary
+            result = self.agent.run(clean_prompt)
+            # Ensure we return a clean string
+            return str(result).strip()
+        except Exception as e:
+            print(f"Error during agent execution: {e}")
+            return "Error solving question"
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    space_id = os.getenv("SPACE_ID")
+
+    if not profile:
+        return "Please Login to Hugging Face with the button.", None
+
+    username = f"{profile.username}"
+    api_url = DEFAULT_API_URL
+    questions_url = f"{api_url}/questions"
+    submit_url = f"{api_url}/submit"
+
+    # 1. Instantiate Agent
+    try:
+        agent = BasicAgent()
+    except Exception as e:
+        return f"Error initializing agent: {e}", None
+    
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
+
+    # 2. Fetch Questions
+    try:
+        response = requests.get(questions_url, timeout=15)
+        response.raise_for_status()
+        questions_data = response.json()
+    except Exception as e:
+        return f"Error fetching questions: {e}", None
+
+    # 3. Run Agent on Questions
+    results_log = []
+    answers_payload = []
+    
+    for item in questions_data:
+        task_id = item.get("task_id")
+        question_text = item.get("question")
+        
+        if not task_id or question_text is None:
+            continue
+
+        # Optional: Handle file downloads for specific tasks
+        # You could add logic here to download from f"{api_url}/files/{task_id}"
+
+        try:
+            submitted_answer = agent(question_text)
+            answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
+        except Exception as e:
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"ERROR: {e}"})
+
+    # 4. Submit Results
+    submission_data = {
+        "username": username.strip(), 
+        "agent_code": agent_code, 
+        "answers": answers_payload
+    }
+
+    try:
+        response = requests.post(submit_url, json=submission_data, timeout=60)
+        response.raise_for_status()
+        result_data = response.json()
+        
+        final_status = (
+            f"Submission Successful!\n"
+            f"User: {result_data.get('username')}\n"
+            f"Score: {result_data.get('score')}% "
+            f"({result_data.get('correct_count')}/{result_data.get('total_attempted')} correct)"
+        )
+        return final_status, pd.DataFrame(results_log)
+    except Exception as e:
+        return f"Submission Failed: {e}", pd.DataFrame(results_log)
+
+# --- Gradio Interface ---
+with gr.Blocks() as demo:
+    gr.Markdown("# GAIA Agent Evaluation Runner")
+    gr.Markdown("This agent uses `smolagents` with a Python interpreter and Web Search to solve tasks.")
+
+    gr.LoginButton()
+    run_button = gr.Button("Run Evaluation & Submit All Answers", variant="primary")
+
+    status_output = gr.Textbox(label="Status", lines=4)
+    results_table = gr.DataFrame(label="Detailed Results", wrap=True)
+
+    run_button.click(
+        fn=run_and_submit_all,
+        outputs=[status_output, results_table]
+    )
+
+if __name__ == "__main__":
+    demo.launch()