import os import gradio as gr import requests import pandas as pd import logging import time import traceback from typing import Dict, Any, Optional, Tuple, List, Union # (Keep Constants as is) # --- Constants --- DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" # --- Import our GAIA Agent --- from gaia_agent import GAIAAgent # Configure logging logging.basicConfig( level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s' ) logger = logging.getLogger("gaia_evaluation") # Initialize the GAIA Agent def initialize_agent(): try: agent = GAIAAgent() logger.info("GAIA Agent initialized successfully") return agent except Exception as e: logger.error(f"Error initializing GAIA Agent: {e}") logger.error(traceback.format_exc()) return None # Important: Match the exact signature from the template def run_and_submit_all(profile=None): """ Fetches all questions, runs the GAIA Agent on them, submits all answers, and displays the results. """ # --- Determine HF Space Runtime URL and Repo URL --- space_id = os.getenv("SPACE_ID", "") # Check if user is signed in if profile: if hasattr(profile, 'username'): username = profile.username logger.info(f"User logged in: {username}") else: username = str(profile) logger.info(f"Using provided username: {username}") else: logger.warning("User not logged in.") return "Please Login to Hugging Face with the button.", None api_url = DEFAULT_API_URL questions_url = f"{api_url}/questions" submit_url = f"{api_url}/submit" # 1. Instantiate Agent try: agent = initialize_agent() if agent is None: error_msg = "Error initializing GAIA Agent. Check logs for details." logger.error(error_msg) return error_msg, None except Exception as e: error_msg = f"Error instantiating agent: {e}" logger.error(error_msg) return error_msg, None # Link to the code repository in Hugging Face Space agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" logger.info(f"Using agent code URL: {agent_code}") # 2. Fetch Questions logger.info(f"Fetching questions from: {questions_url}") try: response = requests.get(questions_url, timeout=15) response.raise_for_status() questions_data = response.json() if not questions_data: logger.warning("Fetched questions list is empty.") return "Fetched questions list is empty or invalid format.", None logger.info(f"Fetched {len(questions_data)} questions.") except requests.exceptions.RequestException as e: error_msg = f"Error fetching questions: {e}" logger.error(error_msg) return error_msg, None except requests.exceptions.JSONDecodeError as e: error_msg = f"Error decoding JSON response from questions endpoint: {e}" logger.error(f"{error_msg}\nResponse text: {response.text[:500]}") return error_msg, None except Exception as e: error_msg = f"An unexpected error occurred fetching questions: {e}" logger.error(error_msg) logger.error(traceback.format_exc()) return error_msg, None # 3. Run the GAIA Agent on all questions results_log = [] answers_payload = [] logger.info(f"Running agent on {len(questions_data)} questions...") for i, item in enumerate(questions_data): # Extract question information task_id = item.get("task_id", item.get("id", f"q{i+1}")) question_text = item.get("question") if not task_id or question_text is None: logger.warning(f"Skipping item with missing task_id or question: {item}") continue logger.info(f"Processing question {i+1}/{len(questions_data)}: {question_text[:50]}...") start_time = time.time() try: # Process the question with the GAIA Agent submitted_answer = agent.process_question(question_text) processing_time = time.time() - start_time # Add to submission payload answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer}) # Store for results display results_log.append({ "Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer, "Processing Time": f"{processing_time:.2f}s", "Status": "Success" }) logger.info(f"Question {i+1} processed successfully in {processing_time:.2f}s") except Exception as e: error_msg = f"Error running agent on task {task_id}: {e}" logger.error(error_msg) logger.error(traceback.format_exc()) results_log.append({ "Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}", "Processing Time": f"{time.time() - start_time:.2f}s", "Status": "Error" }) # Check if we have answers to submit if not answers_payload: logger.warning("Agent did not produce any answers to submit.") return "Agent did not produce any answers to submit.", pd.DataFrame(results_log) # 4. Prepare Submission submission_data = { "username": username.strip(), "agent_code": agent_code, "answers": answers_payload } status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..." logger.info(status_update) # 5. Submit logger.info(f"Submitting {len(answers_payload)} answers to: {submit_url}") try: response = requests.post(submit_url, json=submission_data, timeout=60) response.raise_for_status() result_data = response.json() # Extract results data correct_count = result_data.get("correct_count", 0) total_attempted = result_data.get("total_attempted", 0) score = result_data.get("score", "N/A") final_status = ( f"Submission Successful!\n" f"User: {result_data.get('username', username)}\n" f"Overall Score: {score}% " f"({correct_count}/{total_attempted} correct)\n" f"Message: {result_data.get('message', 'No message received.')}" ) logger.info("Submission successful.") results_df = pd.DataFrame(results_log) return final_status, results_df except requests.exceptions.HTTPError as e: error_detail = f"Server responded with status {e.response.status_code}." try: error_json = e.response.json() error_detail += f" Detail: {error_json.get('detail', e.response.text)}" except: error_detail += f" Response: {e.response.text[:500]}" status_message = f"Submission Failed: {error_detail}" logger.error(status_message) results_df = pd.DataFrame(results_log) return status_message, results_df except requests.exceptions.Timeout: status_message = "Submission Failed: The request timed out." logger.error(status_message) results_df = pd.DataFrame(results_log) return status_message, results_df except requests.exceptions.RequestException as e: status_message = f"Submission Failed: Network error - {e}" logger.error(status_message) results_df = pd.DataFrame(results_log) return status_message, results_df except Exception as e: status_message = f"An unexpected error occurred during submission: {e}" logger.error(status_message) logger.error(traceback.format_exc()) results_df = pd.DataFrame(results_log) return status_message, results_df # --- Build Gradio Interface using Blocks --- with gr.Blocks() as demo: gr.Markdown("# GAIA Agent Evaluation Runner") gr.Markdown( """ **Instructions:** 1. Log in to your Hugging Face account using the button below. This uses your HF username for submission. 2. Click 'Run Evaluation & Submit All Answers' to fetch questions, run the GAIA agent, submit answers, and see the score. --- **Note:** Running the evaluation may take some time as the agent processes all questions. Please be patient. """ ) # Create login button login_btn = gr.LoginButton() # Output text and result areas status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False) results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True) # Run button run_button = gr.Button("Run Evaluation & Submit All Answers") # Set up event handler (simplified) run_button.click( fn=run_and_submit_all, inputs=login_btn, # Connect login button directly outputs=[status_output, results_table] ) if __name__ == "__main__": print("\n" + "-"*30 + " App Starting " + "-"*30) # Check for SPACE_HOST and SPACE_ID at startup for information space_host_startup = os.getenv("SPACE_HOST") space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup if space_host_startup: print(f"✅ SPACE_HOST found: {space_host_startup}") print(f" Runtime URL should be: https://{space_host_startup}.hf.space") else: print("ℹ️ SPACE_HOST environment variable not found (running locally?).") if space_id_startup: # Print repo URLs if SPACE_ID is found print(f"✅ SPACE_ID found: {space_id_startup}") print(f" Repo URL: https://huggingface.co/spaces/{space_id_startup}") print(f" Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main") else: print("ℹ️ SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.") print("-"*(60 + len(" App Starting ")) + "\n") print("Launching Gradio Interface for GAIA Agent Evaluation...") # Determine launch parameters based on environment is_running_in_space = bool(space_host_startup and space_id_startup) if is_running_in_space: # Production settings for Hugging Face Space demo.launch( debug=False, share=False, server_name="0.0.0.0" ) else: # Development settings for local testing demo.launch( debug=True, share=False )