import os import gradio as gr import requests import inspect import pandas as pd import ast import operator import time import json from datetime import datetime from typing import List, Dict, Any, Optional, Annotated from langgraph.graph import Graph, StateGraph from langgraph.prebuilt import ToolNode from tools import simple_search from openai import OpenAI from typing_extensions import TypedDict def override(_, new): return new def merge_dicts(dict1: Dict, dict2: Dict) -> Dict: """Merge two dictionaries, with values from dict2 taking precedence.""" return {**dict1, **dict2} print("trial") # (Keep Constants as is) # --- Constants --- DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" OPENAI_API_KEY = os.getenv("OPENAI_API_KEY") # Make sure to set this environment variable # Create logs directory if it doesn't exist LOGS_DIR = "question_logs" os.makedirs(LOGS_DIR, exist_ok=True) def log_to_file(task_id: str, question: str, log_data: Dict[str, Any]): """Store logs for a question in a JSON file.""" timestamp = datetime.now().strftime("%Y%m%d_%H%M%S") filename = f"{LOGS_DIR}/question_{task_id}_{timestamp}.json" log_entry = { "task_id": task_id, "question": question, "timestamp": timestamp, "logs": log_data } with open(filename, 'w', encoding='utf-8') as f: json.dump(log_entry, f, indent=2, ensure_ascii=False) print(f"Logs saved to {filename}") class AgentState(TypedDict): question: Annotated[str, override] current_step: Annotated[str, override] tool_output: Annotated[str, override] final_answer: Annotated[str, override] history: Annotated[List[Dict[str, str]], operator.add] needs_more_info: Annotated[bool, override] search_query: Annotated[str, override] task_id: Annotated[str, override] logs: Annotated[Dict[str, Any], merge_dicts] # Use merge_dicts instead of operator.add class BasicAgent: def __init__(self): print("Initializing BasicAgent with OpenAI...") if not OPENAI_API_KEY: raise ValueError("OPENAI_API_KEY environment variable not set. Please set your OpenAI API key.") # Initialize OpenAI client self.llm = OpenAI(api_key=OPENAI_API_KEY) # Create the agent workflow print("Creating workflow variable") self.workflow = self._create_workflow() print("BasicAgent initialization complete.") def _call_llm_api(self, prompt: str) -> str: """Call the model and return the raw text output.""" try: print("=== Sending prompt ===") print(prompt[:500]) response = self.llm.chat.completions.create( model="gpt-4.1-nano", messages=[ {"role": "system", "content": "You are a helpful AI assistant that provides clear and concise answers."}, {"role": "user", "content": prompt} ], max_tokens=200, temperature=0.7, top_p=0.95, frequency_penalty=0.1 ) print("=== Received response ===") response_text = response.choices[0].message.content print(response_text) return response_text except Exception as e: print(f"Error calling LLM API: {e}") return f"Error getting response from LLM: {str(e)}" def _create_workflow(self) -> Graph: """Create the agent workflow using LangGraph.""" # Create the workflow with state schema print("Creating Stategraph : error happens here?") workflow = StateGraph(state_schema=AgentState) print("Stategraph created") # Add nodes workflow.add_node("analyze", self._analyze_question) workflow.add_node("search", self._use_search) workflow.add_node("generate_answer", self._generate_final_answer) # Define edges workflow.add_edge("analyze", "search") workflow.add_edge("analyze", "generate_answer") workflow.add_edge("search", "generate_answer") # Define conditional edges def router(state: AgentState) -> str: if state["current_step"] == 'search': return 'search' elif state["current_step"] == 'final_answer': return 'generate_answer' return 'analyze' workflow.add_conditional_edges( "analyze", router, { "search": "search", "final_answer": "generate_answer" } ) # Set entry and exit points workflow.set_entry_point("analyze") workflow.set_finish_point("generate_answer") return workflow.compile() def _analyze_question(self, state: AgentState) -> AgentState: """Analyze the question and determine the next step.""" prompt = f"""Analyze this question and determine what needs to be done: {state['question']} Return ONLY a Python dictionary in this exact format, with no other text or explanation: {{ "needs_search": true/false, "search_query": "query if needed" }}""" try: llm_response = self._call_llm_api(prompt) print("\n=== Analyze Question LLM Response ===") print(f"Input: {state['question']}") print(f"LLM Response: {llm_response}") # Log the analysis step state["logs"]["analyze"] = { "prompt": prompt, "response": llm_response, "timestamp": datetime.now().isoformat() } analysis = ast.literal_eval(llm_response) state["needs_more_info"] = analysis.get('needs_search', False) state["search_query"] = analysis.get('search_query', '') if analysis.get('needs_search', False): state["current_step"] = 'search' else: state["current_step"] = 'final_answer' except (ValueError, SyntaxError) as e: print(f"Error parsing LLM response: {e}") # Default to search if we can't parse the response state["needs_more_info"] = True state["search_query"] = state["question"] state["current_step"] = 'search' # Log the error state["logs"]["analyze_error"] = { "error": str(e), "timestamp": datetime.now().isoformat() } return state def _use_search(self, state: AgentState) -> AgentState: """Use the search tool.""" time.sleep(2) # Sleep before search try: print("\n=== Search Tool ===") print(f"Search Query: {state['search_query']}") # Use the simplified search function search_results = simple_search( query=state["search_query"], max_results=3 ) print("Search Results:") for i, result in enumerate(search_results, 1): print(f"{i}. {result}") # Log the search step state["logs"]["search"] = { "query": state["search_query"], "results": search_results, "timestamp": datetime.now().isoformat() } state["history"].append({ 'step': 'search', 'query': state["search_query"], 'results': search_results }) state["needs_more_info"] = False state["current_step"] = 'final_answer' except Exception as e: print(f"Search Error: {e}") state["history"].append({ 'step': 'search_error', 'error': str(e) }) state["current_step"] = 'final_answer' # Log the error state["logs"]["search_error"] = { "error": str(e), "timestamp": datetime.now().isoformat() } return state def _generate_final_answer(self, state: AgentState) -> AgentState: """Generate the final answer based on all gathered information.""" history_str = "\n".join([f"{h['step']}: {h.get('output', h.get('results', h.get('error', '')))}" for h in state["history"]]) prompt = f"""Question: {state['question']} History of steps taken: {history_str} Return ONLY the direct answer to the question. Do not include any explanations, introductions, or formatting. Just the answer.""" print("\n=== Generate Final Answer ===") print(f"Question: {state['question']}") print("History:") print(history_str) llm_response = self._call_llm_api(prompt) print("\nFinal Answer:") print(llm_response) # Log the final answer generation state["logs"]["final_answer"] = { "prompt": prompt, "response": llm_response, "history": history_str, "timestamp": datetime.now().isoformat() } state["final_answer"] = llm_response return state def __call__(self, question: str, task_id: str = "unknown") -> str: """Process a question through the agent workflow.""" print(f"Agent received question: {question[:50]}...") try: # Initialize the state initial_state: AgentState = { "question": question, "current_step": "analyze", "tool_output": "", "final_answer": "", "history": [], "needs_more_info": False, "search_query": "", "task_id": task_id, "logs": {} } # Run the workflow final_state = self.workflow.invoke(initial_state) # Save logs to file log_to_file( task_id=final_state["task_id"], question=final_state["question"], log_data=final_state["logs"] ) return final_state["final_answer"] except Exception as e: print(f"Error in agent processing: {e}") return f"I encountered an error while processing your question: {str(e)}" def run_and_submit_all(profile: gr.OAuthProfile | None): """ Fetches all questions, runs the BasicAgent on them, 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 print("Space ID: ", 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 Agent try: print("Initializing agent: trial ") agent = BasicAgent() print("Agent initialized successfully with workflow.") 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(f"Agent code location: {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 # 3. Run your Agent results_log = [] answers_payload = [] print(f"Running agent workflow 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: # Initialize the state for this question initial_state = { "question": question_text, "current_step": "analyze", "tool_output": "", "final_answer": "", "history": [], "needs_more_info": False, "search_query": "", "task_id": task_id, "logs": {} } # Run the workflow for this question print(f"\nProcessing question {task_id}: {question_text[:50]}...") final_state = agent.workflow.invoke(initial_state) # Log the workflow history workflow_history = "\n".join([ f"Step: {h['step']}\n" + f"Input: {h.get('input', h.get('query', ''))}\n" + f"Output: {h.get('output', h.get('results', h.get('error', '')))}" for h in final_state["history"] ]) # Add to results submitted_answer = final_state["final_answer"] 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, "Workflow History": workflow_history }) print(f"Completed question {task_id} with {len(final_state['history'])} workflow steps") except Exception as e: print(f"Error running agent workflow on task {task_id}: {e}") results_log.append({ "Task ID": task_id, "Question": question_text, "Submitted Answer": f"WORKFLOW ERROR: {e}", "Workflow History": "Error occurred before workflow completion" }) 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 workflow 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 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)