Spaces:
Sleeping
Sleeping
| import os | |
| import gradio as gr | |
| import requests | |
| import pandas as pd | |
| from agent import create_agent, fetch_random_question | |
| # --- Constants --- | |
| DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" | |
| def run_and_submit_all(profile): | |
| """ | |
| Fetch all questions, run the SmolAgent on them, submit all answers, | |
| and display the results. | |
| """ | |
| space_id = os.getenv("SPACE_ID") | |
| if profile is None: | |
| return "Please login to Hugging Face with the button.", None | |
| username = profile.username | |
| # Instantiate the agent | |
| try: | |
| agent = create_agent() | |
| except Exception as e: | |
| return f"Error initializing agent: {e}", None | |
| agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" | |
| # Fetch questions | |
| try: | |
| response = requests.get(f"{DEFAULT_API_URL}/questions", timeout=15) | |
| response.raise_for_status() | |
| questions = response.json() | |
| except Exception as e: | |
| return f"Error fetching questions: {e}", None | |
| if not questions: | |
| return "No questions fetched.", None | |
| # Run agent on each question | |
| results = [] | |
| payload = [] | |
| for q in questions: | |
| task_id = q.get("task_id") | |
| question_text = q.get("question") | |
| if not task_id or not question_text: | |
| continue | |
| try: | |
| answer = agent.run(question=question_text) | |
| except Exception as e: | |
| answer = f"ERROR: {e}" | |
| results.append({"Task ID": task_id, "Question": question_text, "Answer": answer}) | |
| payload.append({"task_id": task_id, "submitted_answer": answer}) | |
| # Submit answers | |
| submit_payload = {"username": username, "agent_code": agent_code, "answers": payload} | |
| try: | |
| resp = requests.post(f"{DEFAULT_API_URL}/submit", json=submit_payload, timeout=60) | |
| resp.raise_for_status() | |
| data = resp.json() | |
| status = ( | |
| f"Submission Successful!\n" | |
| f"User: {data.get('username')}\n" | |
| f"Score: {data.get('score')}% ({data.get('correct_count')}/{data.get('total_attempted')})\n" | |
| f"Message: {data.get('message')}" | |
| ) | |
| except Exception as e: | |
| status = f"Submission Failed: {e}" | |
| return status, pd.DataFrame(results) | |
| def test_random_question(profile): | |
| """ | |
| Fetch a random GAIA question and return its answer by the agent. | |
| """ | |
| if profile is None: | |
| return "Please login to test.", "" | |
| try: | |
| q = fetch_random_question() | |
| agent = create_agent() | |
| ans = agent.run(question=q.get("question", "")) | |
| return q.get("question", ""), ans | |
| except Exception as e: | |
| return f"Error: {e}", "" | |
| # --- Gradio Interface --- | |
| with gr.Blocks() as demo: | |
| gr.Markdown("# SmolAgent Evaluation Runner") | |
| gr.Markdown( | |
| """ | |
| **Istruzioni:** | |
| 1. Clone questo space e definisci la logica in agent.py. | |
| 2. Effettua il login con il tuo account Hugging Face. | |
| 3. Usa 'Run Evaluation & Submit All Answers' o 'Test Random Question'. | |
| """ | |
| ) | |
| # OAuth login and user state | |
| login = gr.LoginButton() | |
| user = gr.State() | |
| # On login, store profile in state | |
| login.click(fn=lambda profile: profile, inputs=[login], outputs=[user]) | |
| run_all = gr.Button("Run Evaluation & Submit All Answers") | |
| test = gr.Button("Test Random Question") | |
| status = gr.Textbox(label="Status / Risultato", lines=5, interactive=False) | |
| table = gr.DataFrame(label="Risultati Completi", wrap=True) | |
| qbox = gr.Textbox(label="Domanda Casuale", lines=3, interactive=False) | |
| abox = gr.Textbox(label="Risposta Agente", lines=3, interactive=False) | |
| # Use stored user state as input | |
| run_all.click(fn=run_and_submit_all, inputs=[user], outputs=[status, table]) | |
| test.click(fn=test_random_question, inputs=[user], outputs=[qbox, abox]) | |
| if __name__ == "__main__": | |
| demo.launch(debug=True, share=False) | |
| demo.launch(debug=True, share=False) | |