from __future__ import annotations import os import gradio as gr import requests import pandas as pd from langchain_openai import ChatOpenAI from langgraph.graph import StateGraph, START, END from langchain.schema import HumanMessage, SystemMessage, AIMessage # Create a ToolNode that knows about your web_search function import json from state import AgentState # --- Constants --- DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" import json from typing import Any, Dict, List, Optional # ─────────────────────────── External tools ────────────────────────────── from tools import ( wikipedia_search_tool, ocr_image_tool, audio_transcriber_tool, parse_excel_tool ) # ─────────────────────────── Configuration ─────────────────────────────── LLM = ChatOpenAI(model_name="gpt-4.1", temperature=0.0) MAX_TOOL_CALLS = 5 # ─────────────────────────── Helper utilities ──────────────────────────── def safe_json(text: str) -> Optional[Dict[str, Any]]: """Parse the *first* mapping‑literal in `text`. • Accepts **strict JSON** or Python‑style single‑quoted dicts. • Ignores markdown fences / leading commentary. """ import re, json, ast # Strip ``` fences if any if text.strip().startswith("```"): text = re.split(r"```+", text.strip(), maxsplit=2)[1] # Find the first {...} brace, start = 0, None for i, ch in enumerate(text): if ch == '{': if brace == 0: start = i brace += 1 elif ch == '}' and brace: brace -= 1 if brace == 0 and start is not None: candidate = text[start:i+1] # First try strict JSON try: return json.loads(candidate) except json.JSONDecodeError: # Fallback: Python literal (handles single quotes) try: obj = ast.literal_eval(candidate) return obj if isinstance(obj, dict) else None except Exception: return None return None # def brief(d: Dict[str, Any]) -> str: # for k in ("wiki_result", "ocr_result", "transcript"): # if k in d: # return f"{k}: {str(d[k])[:160].replace('\n', ' ')}…" # return "(no output)" # ─────────────────────────── Agent state ⬇ ─────────────────────────────── # ───────────────────────────── Nodes ⬇ ─────────────────────────────────── def tool_selector(state: AgentState) -> AgentState: """Ask the LLM what to do next (wiki / ocr / audio / excel / final).""" if state.tool_calls >= MAX_TOOL_CALLS: state.add(SystemMessage(content="You have reached the maximum number of tool calls. Use the already gathered information to answer the question.")) state.next_action = "final" return state prompt = SystemMessage( content=( "Reply with ONE JSON only (no markdown). Choices:\n" " {'action':'wiki','query':'…'}\n" " {'action':'ocr'}\n" " {'action':'audio'}\n" " {'action':'excel'}\n" " {'action':'final'}\n" "if the tool you want isnt listed above, return {'action':'final'} \n" "Use wiki if you need to search online for information. Keep the query short and concise and accurate. The query should not be a prompt but instad you should search for the relevant information rather than asking for the answer directly.\n" "If the question is about any image, you have to use ocr tool. It will tell you about the image also\n" "Use audio if the question is about an audio file\n" "Use excel if the question is about an excel file\n" ) ) raw = LLM(state.messages + [prompt]).content.strip() print(f"Tool selector response: {raw}") state.add(AIMessage(content=raw)) parsed = safe_json(raw) # parsed = json.loads(raw) # print("parsed : ", parsed) # print(f"Parsed: {parsed}, type: {type(parsed)}") if not parsed or "action" not in parsed: state.next_action = "final" return state # print("reached here") state.next_action = parsed["action"] state.query = parsed.get("query") return state # ------------- tool adapters ------------- def wiki_tool(state: AgentState) -> AgentState: out = wikipedia_search_tool({"wiki_query": state.query or ""}) state.tool_calls += 1 state.add(SystemMessage(content=f"WIKI_TOOL_OUT: {out}")) state.next_action = None return state def ocr_tool(state: AgentState) -> AgentState: out = ocr_image_tool({"task_id": state.task_id, "ocr_path": ""}) state.tool_calls += 1 state.add(SystemMessage(content=f"OCR_TOOL_OUT: {out}")) state.next_action = None return state def audio_tool(state: AgentState) -> AgentState: out = audio_transcriber_tool({"task_id": state.task_id, "audio_path": ""}) state.tool_calls += 1 state.add(SystemMessage(content=f"AUDIO_TOOL_OUT: {out}")) state.next_action = None return state def excel_tool(state: AgentState) -> AgentState: result = parse_excel_tool({ "task_id": state.task_id, "excel_sheet_name": "" }) out = {"excel_result": result} state.tool_calls += 1 state.add(SystemMessage(content=f"EXCEL_TOOL_OUT: {out}")) state.next_action = None return state # ------------- final answer ------------- def final_node(state: AgentState) -> AgentState: print("reached final node") wrap = SystemMessage( content="Using everything so far, reply ONLY with {'final_answer':'…'}. YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string. \n" "reply **only** with " "{\"final_answer\":\"…\"} (no markdown, no commentary)." ) raw = LLM(state.messages + [wrap]).content.strip() # print("raw : ", raw) state.add(AIMessage(content=raw)) parsed = safe_json(raw) # print("parsed : ", parsed, "type : ", type(parsed)) state.final_answer = parsed.get("final_answer") if parsed else "Unable to parse final answer." # print("state.final_answer : ", state.final_answer) return state # ─────────────────────────── Graph wiring ─────────────────────────────── graph = StateGraph(AgentState) # Register nodes for name, fn in [ ("tool_selector", tool_selector), ("wiki_tool", wiki_tool), ("ocr_tool", ocr_tool), ("audio_tool", audio_tool), ("excel_tool", excel_tool), ("final_node", final_node), ]: graph.add_node(name, fn) # Edges graph.add_edge(START, "tool_selector") def dispatch(state: AgentState) -> str: return { "wiki": "wiki_tool", "ocr": "ocr_tool", "audio": "audio_tool", "excel": "excel_tool", "final": "final_node", }.get(state.next_action, "final_node") graph.add_conditional_edges( "tool_selector", dispatch, { "wiki_tool": "wiki_tool", "ocr_tool": "ocr_tool", "audio_tool": "audio_tool", "excel_tool": "excel_tool", "final_node": "final_node", }, ) # tools loop back to selector for tool_name in ("wiki_tool", "ocr_tool", "audio_tool", "excel_tool"): graph.add_edge(tool_name, "tool_selector") # final_answer → END graph.add_edge("final_node", END) compiled_graph = graph.compile() # ─────────────────────────── Public API ──────────────────────────────── def answer(question: str, task_id: Optional[str] = None) -> str: """Run the agent and return whatever FINAL_ANSWER the graph produces.""" init_state = AgentState(question, task_id) init_state.add(SystemMessage(content="You are a helpful assistant.")) init_state.add(HumanMessage(content=question)) # IMPORTANT: invoke() returns a **new** state instance (or an AddableValuesDict), # not the object we pass in. Use the returned value to fetch final_answer. out_state = compiled_graph.invoke(init_state) if isinstance(out_state, dict): # AddableValuesDict behaves like a dict return out_state.get("final_answer", "No answer.") else: # If future versions return the dataclass return getattr(out_state, "final_answer", "No answer.") class BasicAgent: def __init__(self): print("BasicAgent initialized.") def __call__(self, question: str, task_id) -> str: # print(f"Agent received question (first 50 chars): {question[:50]}...") # fixed_answer = "This is a default answer." # print(f"Agent returning fixed answer: {fixed_answer}") print() print() print() print() print(f"Agent received question: {question}") print() return answer(question, task_id) # return fixed_answer 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 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 ( usefull for others so please keep it public) 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 # 3. Run your 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, task_id) 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 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("LangGraph version:", langgraph.__version__) 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 # import langgraph # print("▶︎ LangGraph version:", langgraph.__version__) 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)