""" Basic Agent Evaluation Runner""" import os import gradio as gr import requests import pandas as pd import concurrent.futures import os import gradio as gr import requests import pandas as pd import os from agents import get_manager_agent from typing import Optional import os import time import requests from typing import Optional from pathlib import Path # File cache to avoid repeated downloads FILE_CACHE_DIR = Path("./cache") FILE_CACHE_DIR.mkdir(exist_ok=True, parents=True) DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" class FileCache: """Handles caching of files to avoid repeated downloads.""" @staticmethod def get_cached_path(file_name: str) -> Optional[Path]: """Check if file is already cached.""" cache_path = FILE_CACHE_DIR / file_name return cache_path if cache_path.exists() else None @staticmethod def cache_file(file_name: str, content: bytes) -> Path: """Cache file content.""" cache_path = FILE_CACHE_DIR / file_name with open(cache_path, 'wb') as f: f.write(content) return cache_path @staticmethod def get_file_extension(filename: str) -> str: """Extract file extension from filename.""" return Path(filename).suffix class FileManager: """Handles file retrieval and caching.""" @staticmethod def get_file_by_task_id(task_id: str, file_name: str) -> Optional[Path]: """ Fetch file associated with task ID and cache it. Args: task_id: The ID of the task file_name: The name of the file Returns: Path to the cached file, or None if no file exists """ # Check cache first cached_path = FileCache.get_cached_path(file_name) if cached_path: return cached_path # If not cached, download url = f"{DEFAULT_API_URL}/files/{task_id}" try: response = requests.get(url, timeout=10) response.raise_for_status() # Cache the file content = response.content cached_path = FileCache.cache_file(file_name, content) return cached_path except Exception as e: print(f"Error fetching file for task {task_id}: {e}") return None # (Keep Constants as is) # --- Constants --- class BasicAgent: def __init__(self): print("BasicAgent initialized.") self.agent = get_manager_agent() self.verbose = True def __call__(self, task_id: str, question: str, file_name: str = None) -> str: """ Process a GAIA benchmark question and return the answer Args: question: The question to answer task_file_path: Optional path to a file associated with the question Returns: The answer to the question """ task_file_path = None if file_name: task_file_path = FileManager.get_file_by_task_id(task_id, file_name) try: if self.verbose: print(f"Processing question: {question}") if task_file_path: print(f"With associated file: {task_file_path}") # Create a context with file information if available context = question # If there's a file, read it and include its content in the context if task_file_path: context = f""" Question: {question} This question has an associated file. File is already downloaded at: {task_file_path} Analyze the file content above to answer the question using tools. """ # Check for special cases that need specific formatting # Reversed text questions if question.startswith(".") or ".rewsna eht sa" in question: context = f""" This question appears to be in reversed text. Here's the reversed version: {question[::-1]} Now answer the question above. Remember to format your answer exactly as requested. """ # Add a prompt to ensure precise answers full_prompt = f"""{context} When answering, provide ONLY the precise answer requested. Do not include explanations, steps, reasoning, or additional text. Be direct and specific. GAIA benchmark requires exact matching answers. For example, if asked "What is the capital of France?", respond simply with "Paris". """ # Run the agent with the question answer = self.agent.run(full_prompt) # Clean up the answer to ensure it's in the expected format # Remove common prefixes that models often add answer = self._clean_answer(answer) if self.verbose: print(f"Generated answer: {answer}") return answer except Exception as e: error_msg = f"Error answering question: {e}" if self.verbose: print(error_msg) return error_msg def _clean_answer(self, answer: any) -> str: """ Clean up the answer to remove common prefixes and formatting that models often add but that can cause exact match failures. Args: answer: The raw answer from the model Returns: The cleaned answer as a string """ # Convert non-string types to strings if not isinstance(answer, str): # Handle numeric types (float, int) if isinstance(answer, float): # Format floating point numbers properly # Check if it's an integer value in float form (e.g., 12.0) if answer.is_integer(): formatted_answer = str(int(answer)) else: # For currency values that might need formatting if abs(answer) >= 1000: formatted_answer = f"${answer:,.2f}" else: formatted_answer = str(answer) return formatted_answer elif isinstance(answer, int): return str(answer) else: # For any other type return str(answer) # Now we know answer is a string, so we can safely use string methods # Normalize whitespace answer = answer.strip() # Remove common prefixes and formatting that models add prefixes_to_remove = [ "The answer is ", "Answer: ", "Final answer: ", "The result is ", "To answer this question: ", "Based on the information provided, ", "According to the information: ", ] for prefix in prefixes_to_remove: if answer.startswith(prefix): answer = answer[len(prefix):].strip() # Remove quotes if they wrap the entire answer if (answer.startswith('"') and answer.endswith('"')) or (answer.startswith("'") and answer.endswith("'")): answer = answer[1:-1].strip() return answer def run_and_submit_all(profile: gr.OAuthProfile | None): """ Fetches all questions, runs the BasicAgent on them in parallel (up to 10 at a time), submits all answers, and displays the results. """ # --- Determine HF Space Runtime URL and Repo URL --- space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code if profile: username= f"{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 (modify this part to create your agent) try: agent = BasicAgent() except Exception as e: print(f"Error instantiating agent: {e}") return f"Error initializing agent: {e}", None # In the case of an app running as a hugging Face space, this link points toward your codebase agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" print(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 requests.exceptions.JSONDecodeError as e: print(f"Error decoding JSON response from questions endpoint: {e}") print(f"Response text: {response.text[:500]}") return f"Error decoding server response for 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 # Helper function to process a single question def process_question(item): task_id = item.get("task_id") question_text = item.get("question") file_name = item.get("file_name") if not task_id or question_text is None: print(f"Skipping item with missing task_id or question: {item}") return None try: submitted_answer = BasicAgent()(task_id, question_text, file_name) return { "task_id": task_id, "question": question_text, "submitted_answer": submitted_answer, "error": None } except Exception as e: # raise e print(f"Error running agent on task {task_id}: {e}") return { "task_id": task_id, "question": question_text, "submitted_answer": f"AGENT ERROR: {e}", "error": str(e) } # 3. Run your Agent in parallel (up to 10 at a time) results_log = [] answers_payload = [] print(f"Running agent on {len(questions_data)} questions with up to 10 in parallel...") with concurrent.futures.ThreadPoolExecutor(max_workers=19) as executor: # Submit all questions to the thread pool future_to_item = { executor.submit(process_question, item): item for item in questions_data } # Collect results as they complete for future in concurrent.futures.as_completed(future_to_item): result = future.result() if result is not None: if result["error"] is None: answers_payload.append({ "task_id": result["task_id"], "submitted_answer": result["submitted_answer"] }) results_log.append({ "Task ID": result["task_id"], "Question": result["question"], "Submitted Answer": result["submitted_answer"] }) 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) # save to csv results_df.to_csv("results.csv", index=False) 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 requests.exceptions.Timeout: status_message = "Submission Failed: The request timed out." print(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}" 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 # --- Build Gradio Interface using Blocks --- with gr.Blocks() as demo: gr.Markdown("# Basic Agent Evaluation Runner") gr.Markdown( """ **Instructions:** 1. Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ... 2. Log in to your Hugging Face account using the button below. This uses your HF username for submission. 3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score. --- **Disclaimers:** Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions). This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async. """ ) gr.LoginButton() run_button = gr.Button("Run Evaluation & Submit All Answers") status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False) # Removed max_rows=10 from DataFrame constructor results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True) run_button.click( fn=run_and_submit_all, 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 Basic Agent Evaluation...") demo.launch(debug=True, share=False)