Spaces:
Sleeping
Sleeping
| """ Working Multi-LLM Agent Evaluation Runner""" | |
| import os | |
| import gradio as gr | |
| import requests | |
| import pandas as pd | |
| from langchain_core.messages import HumanMessage | |
| # Import from veryfinal.py | |
| from veryfinal import UnifiedAgnoEnhancedSystem | |
| # --- Constants --- | |
| DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" | |
| # --- Working Agent Definition --- | |
| class WorkingMultiLLMAgent: | |
| """A working multi-LLM agent that actually answers questions""" | |
| def __init__(self): | |
| print("Working Multi-LLM Agent initialized.") | |
| try: | |
| self.system = UnifiedAgnoEnhancedSystem() | |
| print("✅ Working system built successfully.") | |
| except Exception as e: | |
| print(f"❌ Error building system: {e}") | |
| self.system = None | |
| def __call__(self, question: str) -> str: | |
| print(f"Processing: {question[:100]}...") | |
| if self.system is None: | |
| return "Error: System not initialized" | |
| try: | |
| answer = self.system.process_query(question) | |
| # Validation | |
| if not answer or answer == question or len(answer.strip()) == 0: | |
| return "Information not available" | |
| return answer.strip() | |
| except Exception as e: | |
| return f"Error: {str(e)}" | |
| def run_and_submit_all(profile: gr.OAuthProfile | None): | |
| """Run evaluation with working agent""" | |
| space_id = os.getenv("SPACE_ID") | |
| 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 Working Agent | |
| try: | |
| agent = WorkingMultiLLMAgent() | |
| if agent.system is None: | |
| return "Error: Failed to initialize working agent", None | |
| except Exception as e: | |
| return f"Error initializing agent: {e}", None | |
| agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" if space_id else "No space ID" | |
| # 2. Fetch Questions | |
| try: | |
| response = requests.get(questions_url, timeout=15) | |
| response.raise_for_status() | |
| questions_data = response.json() | |
| if not questions_data: | |
| return "No questions fetched", None | |
| print(f"✅ Fetched {len(questions_data)} questions") | |
| except Exception as e: | |
| return f"Error fetching questions: {e}", None | |
| # 3. Process Questions | |
| results_log = [] | |
| answers_payload = [] | |
| for i, item in enumerate(questions_data): | |
| task_id = item.get("task_id") | |
| question_text = item.get("question") | |
| if not task_id or question_text is None: | |
| continue | |
| print(f"Processing {i+1}/{len(questions_data)}: {task_id}") | |
| try: | |
| answer = agent(question_text) | |
| # Prevent question repetition | |
| if answer == question_text or answer.startswith(question_text): | |
| answer = "Information not available" | |
| answers_payload.append({"task_id": task_id, "submitted_answer": answer}) | |
| results_log.append({ | |
| "Task ID": task_id, | |
| "Question": question_text[:100] + "..." if len(question_text) > 100 else question_text, | |
| "Submitted Answer": answer[:200] + "..." if len(answer) > 200 else answer | |
| }) | |
| except Exception as e: | |
| error_msg = f"ERROR: {e}" | |
| answers_payload.append({"task_id": task_id, "submitted_answer": error_msg}) | |
| results_log.append({ | |
| "Task ID": task_id, | |
| "Question": question_text[:100] + "..." if len(question_text) > 100 else question_text, | |
| "Submitted Answer": error_msg | |
| }) | |
| if not answers_payload: | |
| return "No answers generated", pd.DataFrame(results_log) | |
| # 4. Submit Results | |
| submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload} | |
| 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"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', 'Success')}" | |
| ) | |
| return final_status, pd.DataFrame(results_log) | |
| except Exception as e: | |
| return f"❌ Submission Failed: {e}", pd.DataFrame(results_log) | |
| # --- Gradio Interface --- | |
| with gr.Blocks() as demo: | |
| gr.Markdown("# Working Multi-LLM Agent System") | |
| gr.Markdown( | |
| """ | |
| **✅ This is a WORKING system that will actually answer questions!** | |
| **Features:** | |
| - **Groq Llama-3 70B**: High-quality responses | |
| - **Smart Routing**: Math, search, wiki, and general queries | |
| - **Web Search**: Tavily integration for current information | |
| - **Wikipedia**: Encyclopedic knowledge access | |
| - **Robust Error Handling**: Fallbacks and validation | |
| **Instructions:** | |
| 1. Log in with your Hugging Face account | |
| 2. Click 'Run Evaluation & Submit All Answers' | |
| 3. Wait for processing to complete | |
| 4. View your results and score | |
| **Requirements:** | |
| - GROQ_API_KEY in your environment variables | |
| - TAVILY_API_KEY (optional, for web search) | |
| """ | |
| ) | |
| gr.LoginButton() | |
| run_button = gr.Button("🚀 Run Evaluation & Submit All Answers", variant="primary") | |
| status_output = gr.Textbox(label="Status", lines=5, interactive=False) | |
| results_table = gr.DataFrame(label="Results", wrap=True) | |
| run_button.click( | |
| fn=run_and_submit_all, | |
| outputs=[status_output, results_table] | |
| ) | |
| if __name__ == "__main__": | |
| print("🚀 Starting Working Multi-LLM Agent System") | |
| demo.launch(debug=True, share=False) | |