import os import gradio as gr import requests import pandas as pd from huggingface_hub import InferenceClient from duckduckgo_search import DDGS from datasets import load_dataset import wikipediaapi # Constants DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" HF_TOKEN = os.getenv("HF_TOKEN") # Setup Hugging Face client (advanced model) llm_model_id = "HuggingFaceH4/zephyr-7b-beta" hf_client = InferenceClient(llm_model_id, token=HF_TOKEN) # Wikipedia API setup (corrected user-agent) wiki_api = wikipediaapi.Wikipedia( language='en', user_agent='SmartAgent/1.0 (chockqoteewy@gmail.com)' ) # Load a subset of Wikipedia dataset (adjust as needed) wiki_dataset = load_dataset("wikipedia", "20220301.en", split="train[:10000]", trust_remote_code=True) # Search functions def duckduckgo_search(query): with DDGS() as ddgs: results = [r for r in ddgs.text(query, max_results=3)] return "\n".join([r["body"] for r in results if r.get("body")]) or "No results found." def wikipedia_search(query): page = wiki_api.page(query) return page.summary if page.exists() else "No Wikipedia page found." # Comprehensive Agent class SmartAgent: def __init__(self): pass def __call__(self, question: str) -> str: q_lower = question.lower() if any(term in q_lower for term in ["current", "latest", "2024", "2025", "who is the president", "recent", "live"]): return duckduckgo_search(question) wiki_result = wikipedia_search(question) if "No Wikipedia page found" not in wiki_result: return wiki_result try: resp = hf_client.text_generation(question, max_new_tokens=512) return resp except Exception as e: return f"HF LLM error: {e}" # Submission logic def run_and_submit_all(profile: gr.OAuthProfile | None): space_id = os.getenv("SPACE_ID") if profile: username = profile.username print(f"User logged in: {username}") else: 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" agent = SmartAgent() agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" 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 results_log = [] answers_payload = [] correct_answers = 0 for item in questions_data: task_id = item.get("task_id") question_text = item.get("question") if not task_id or not question_text: continue 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}) if not answers_payload: return "Agent did not produce any answers to submit.", pd.DataFrame(results_log) submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload} 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.')}" ) results_df = pd.DataFrame(results_log) return final_status, results_df except Exception as e: return f"Submission Failed: {e}", pd.DataFrame(results_log) # Gradio Interface with gr.Blocks() as demo: gr.Markdown("# Smart Agent Evaluation Runner") gr.Markdown(""" **Instructions:** 1. Clone this space, define your agent logic, tools, packages, etc. 2. Log in to Hugging Face. 3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score. """) 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]) if __name__ == "__main__": demo.launch(debug=True, share=False)