import os import gradio as gr import requests import inspect import pandas as pd import re from PyPDF2 import PdfReader from io import BytesIO import yaml from smolagents import CodeAgent, InferenceClientModel, DuckDuckGoSearchTool, tool, PromptTemplates, VisitWebpageTool import wikipedia from bs4 import BeautifulSoup import pdfplumber from youtube_transcript_api import YouTubeTranscriptApi from sportsreference.mlb.roster import Player import whisper import sys, io # (Keep Constants as is) # --- Constants --- DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" # Optional: For question analysis import wikipedia @tool def exec_python(code: str) -> str: """ Executes a short Python code snippet in a sandboxed environment and returns the standard output or any runtime error. Args: code (str): A string containing valid Python code. Typically limited to a few lines of computation, function calls, or print statements. Example: 'for i in range(3): print(i ** 2)' Returns: str: The result printed by the code, or an error message if execution fails. If the code runs but produces no output, returns 'None'. Behavior: - Captures all output from `print()` statements. - Overrides the built-in `print` function to redirect output to a buffer. - Does not persist variables between calls (runs in isolated scope). - No access to file I/O, network, or external libraries unless included in code string. Security Notes: - This function does not use `eval()` and limits execution scope to an empty global environment. - However, because it still uses `exec()`, it should not be exposed to untrusted users without additional sandboxing. Raises: None directly, but returns a string with an error message if execution throws an exception. Example: >>> exec_python('for i in range(3): print(i * 2)') '0\n2\n4' >>> exec_python('x = 5 / 0') 'Error: division by zero' """ buffer = io.StringIO() try: exec(code, {}, {"print": lambda *args: buffer.write(" ".join(str(a) for a in args) + "\n")}) return buffer.getvalue().strip() or "None" except Exception as e: return f"Error: {e}" @tool def youtube_transcript(url: str) -> str: """ Fetches and concatenates the subtitles/transcript of a YouTube video. Args: url (str): Full YouTube video URL. Returns: str: Plain-text transcript of the entire video. Raises: ValueError: If no transcript is available or video ID is invalid. """ vid = url.split("v=")[-1] entries = YouTubeTranscriptApi.get_transcript(vid) return " ".join(item["text"] for item in entries) @tool def wiki_lookup(question: str) -> str: """ Looks up relevant content from Wikipedia based on the question. Args: question (str): What to look up. Returns: str: Summary content from the most relevant Wikipedia page. """ try: wikipedia.set_lang("en") results = wikipedia.search(question, results=1) if not results: return "❌ No relevant Wikipedia page found." page = wikipedia.page(results[0]) return page.content[:1000] except Exception as e: return f"❌ Wikipedia lookup failed: {e}" @tool def transcribe_audio(data: bytes) -> str: """ Transcribes raw audio bytes using Whisper. Args: data (bytes): MP3 or WAV file content. Returns: str: Full transcription. """ model = whisper.load_model("base") result = model.transcribe(data) return result["text"].strip() @tool def baseball_stats(player_query: str, season: int, stat: str) -> dict: """ Fetches baseball stats from SportsReference.com. Args: player_query (str): Player name or filter. season (int): Year of season. stat (str): Which stat (“walks”, “at_bats”, etc.). Returns: Dict with requested numbers. """ p = Player(player_query, year=season) return {stat: getattr(p, stat)} @tool def text_reverse(text: str) -> str: """ Reverses the input string character-by-character. Args: text (str): The input text to reverse. Returns: str: The reversed text. """ return text[::-1] @tool def pdf_scraper(pdf_url: str) -> str: """ Downloads and extracts text from a PDF at a given URL. Args: pdf_url (str): Direct link to the PDF. Returns: str: Full extracted text from the PDF. """ response = requests.get(pdf_url) reader = PdfReader(BytesIO(response.content)) text = "\n".join(page.extract_text() or '' for page in reader.pages) return text @tool def parse_excel(data: bytes, sheet_name: str = None) -> str: """ Parses an Excel file from raw bytes and converts it to a CSV-formatted string. This tool reads the specified sheet or the first sheet by default, preserves column headers, and omits row indices in the output. Args: data (bytes): Raw bytes content of the uploaded Excel file (.xlsx, .xls). sheet_name (str, optional): Name or index of the sheet to read. If None, defaults to the first sheet in the workbook. Returns: str: CSV-formatted text of the sheet's data with headers and without indices. Raises: ValueError: If the provided data is not a valid Excel file. XLRDError: If the specified sheet_name does not exist. """ try: df = pd.read_excel(BytesIO(data), sheet_name=sheet_name) except Exception as e: raise ValueError(f"Failed to parse Excel data: {e}") return df.to_csv(index=False) # SmolAgent wrapper class BasicAgent: def __init__(self): print("🔧 SmolAgent is being initialized.") # model = InferenceClientModel(model_id="Qwen/Qwen2.5-Coder-32B-Instruct", provider="together", api_key=os.environ['TOGETHER_API_KEY']) model = InferenceClientModel( model_id='meta-llama/Llama-3.3-70B-Instruct') self.agent = CodeAgent( model=model, tools=[text_reverse, pdf_scraper, DuckDuckGoSearchTool(), wiki_lookup, VisitWebpageTool(),parse_excel, baseball_stats, youtube_transcript, transcribe_audio, exec_python], add_base_tools=True, planning_interval=2 ) def __call__(self, question: str) -> str: print(f"🤖 Agent received question: {question[:80]}...") try: result = self.agent.run(question) print(f"✅ Agent result: {result}") return result except Exception as e: error_message = f"❌ Error during agent execution: {e}" print(error_message) return error_message # --- Function to test random question --- def test_random_question(): try: # Step 1: Fetch random question res = requests.get(f"{DEFAULT_API_URL}/random-question") if res.status_code != 200: return "❌ Failed to fetch random question.", "", "" q = res.json() question = q["question"] task_id = q["task_id"] print(f"🎯 Testing on task_id: {task_id}") # Step 2: Run agent agent = BasicAgent() answer = agent(question) # Step 3: Submit for evaluation payload = {"task_id": task_id, "answer": answer} eval_res = requests.post(f"{DEFAULT_API_URL}/evaluate", json=payload) if eval_res.status_code != 200: return question, answer, "❌ Evaluation failed." score = eval_res.json().get("score", "No score returned") return question, answer, f"✅ Score: {score}" except Exception as e: return "❌ Error occurred.", "", f"⚠️ {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 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) 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() with gr.Row(): question_box = gr.Textbox(label="Random Question", lines=4) answer_box = gr.Textbox(label="Agent Answer", lines=2) result_box = gr.Textbox(label="Evaluation Result", lines=1) test_button = gr.Button("🔁 Answer Random Question & Evaluate") test_button.click(fn=test_random_question, inputs=[], outputs=[question_box, answer_box, result_box]) 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)