import os import gradio as gr import requests import pandas as pd from smolagents import CodeAgent, DuckDuckGoSearchTool, InferenceClientModel from huggingface_hub import login # Constants DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" HF_TOKEN = os.environ.get("HF_TOKEN", "") # Hugging Face token # Login if token is provided if HF_TOKEN: login(token=HF_TOKEN) class GAIACodeAgent: def __init__(self): """Initialize the advanced agent with tools and capabilities""" model = InferenceClientModel() self.agent = CodeAgent( tools=[DuckDuckGoSearchTool()], model=model ) def __call__(self, question: str) -> str: """Process a question and return an answer""" try: print(f"Agent received question: {question[:50]}...") # Improve the prompt to get better accuracy on exact match questions enriched_prompt = ( f"Answer the following question accurately and concisely. " f"Provide a straightforward answer without unnecessary elaboration. " f"The answer will be evaluated for exact match accuracy.\n\n" f"Question: {question}\n\n" f"Answer: " ) # Run the agent with the enriched prompt response = self.agent.run(enriched_prompt) # Clean up response to improve exact match chances cleaned_response = response.strip() print(f"Agent returning answer: {cleaned_response[:50]}...") return cleaned_response except Exception as e: error_msg = f"Error: {str(e)}" print(error_msg) return error_msg def run_and_submit_all(profile: gr.OAuthProfile | None): """ Fetches all questions, runs the GAIACodeAgent on them, submits all answers, and displays the results. """ # Determine HF Space Runtime URL and Repo URL space_id = os.getenv("SPACE_ID") if profile: username = profile.username print(f"User logged in: {username}") else: print("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 = GAIACodeAgent() except Exception as e: print(f"Error instantiating agent: {e}") return f"Error initializing agent: {e}", None agent_code = f"https://huggingface.co./spaces/{space_id}/tree/main" print(f"Agent code URL: {agent_code}") # 2. Fetch Questions print(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: print("Fetched questions list is empty.") return "Fetched questions list is empty or invalid format.", None print(f"Fetched {len(questions_data)} questions.") except requests.exceptions.RequestException as e: print(f"Error fetching questions: {e}") return f"Error fetching questions: {e}", None except Exception as e: print(f"An unexpected error occurred fetching questions: {e}") return f"An unexpected error occurred fetching questions: {e}", None # 3. Run Agent results_log = [] answers_payload = [] print(f"Running agent on {len(questions_data)} questions...") 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: print(f"Skipping item with missing task_id or question: {item}") continue 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: print(f"Error running agent on task {task_id}: {e}") results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"}) if not answers_payload: print("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}'..." print(status_update) # 5. Submit print(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() final_status = ( f"Submission Successful!\n" f"User: {result_data.get('username')}\n" f"Overall Score: {result_data.get('score', 'N/A')}% " f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n" f"Message: {result_data.get('message', 'No message received.')}" ) print("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 requests.exceptions.JSONDecodeError: error_detail += f" Response: {e.response.text[:500]}" status_message = f"Submission Failed: {error_detail}" print(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}" print(status_message) results_df = pd.DataFrame(results_log) return status_message, results_df def query_single_agent(question): """Run agent on a single question for testing""" try: agent = GAIACodeAgent() response = agent(question) return response except Exception as e: return f"Error: {str(e)}" # Build Gradio Interface with gr.Blocks(title="GAIA Code Agent Evaluation") as demo: gr.Markdown("# GAIA Code Agent Evaluation") gr.Markdown( """ This application helps you evaluate a code agent on the GAIA benchmark. ## Instructions: 1. Log in to your Hugging Face account using the button below 2. You can test the agent with a single question in the "Test Agent" tab 3. Use the "Run Evaluation" tab to run the agent on all GAIA questions and submit answers """ ) with gr.Tab("Test Agent"): question_input = gr.Textbox( label="Enter a question", placeholder="How many seconds would it take for a leopard at full speed to run through Pont des Arts?" ) query_button = gr.Button("Get Answer") response_output = gr.Textbox(label="Agent Response", lines=10) query_button.click(query_single_agent, inputs=question_input, outputs=response_output) with gr.Tab("Run Evaluation"): gr.LoginButton() run_button = gr.Button("Run Evaluation & Submit All Answers") 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.click(fn=run_and_submit_all, outputs=[status_output, results_table]) # Start the app if __name__ == "__main__": demo.launch()