Download app.py from mattibuzzo13/Final_Assignment_Template: direct link, hf CLI and curl.
- Browser
- Download file 24.3 kB
-
https://huggingface.co/spaces/mattibuzzo13/Final_Assignment_Template/resolve/main/app.py
- Command line
-
hf download hf://spaces/mattibuzzo13/Final_Assignment_Template/app.py
-
curl -L -o app.py https://huggingface.co/spaces/mattibuzzo13/Final_Assignment_Template/resolve/main/app.py
24.3 kB
| import os | |
| import gradio as gr | |
| import requests | |
| import inspect | |
| import pandas as pd | |
| import re | |
| import math | |
| import json | |
| import unicodedata | |
| from typing import TypedDict, Annotated, Any, List, Optional | |
| from huggingface_hub import InferenceClient | |
| from langchain_core.messages import AnyMessage, HumanMessage, SystemMessage, AIMessage, ToolMessage | |
| from langchain_core.tools import tool | |
| from langchain_community.tools import DuckDuckGoSearchRun | |
| from langchain_community.utilities import WikipediaAPIWrapper | |
| from langgraph.graph import START, StateGraph | |
| from langgraph.graph.message import add_messages | |
| from langgraph.prebuilt import ToolNode, tools_condition | |
| # (Keep Constants as is) | |
| # --- Constants --- | |
| DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" | |
| def web_search(query: str) -> str: | |
| """Search the web using DuckDuckGo for current facts, news, specific data, recent information, and verification. Returns top search results.""" | |
| try: | |
| return DuckDuckGoSearchRun().run(query) | |
| except Exception as e: | |
| return f"Search error: {e}" | |
| def _wikipedia_api_query(title: str) -> str: | |
| """Get plain text extract from English Wikipedia for a page title.""" | |
| import urllib.parse | |
| url = ( | |
| "https://en.wikipedia.org/w/api.php" | |
| "?action=query&format=json&prop=extracts&explaintext=1&titles=" | |
| + urllib.parse.quote(title) | |
| ) | |
| headers = {"User-Agent": "Mozilla/5.0 (compatible; AgentBot/1.0; +https://example.com/bot)"} | |
| r = requests.get(url, timeout=15, headers=headers) | |
| try: | |
| r.raise_for_status() | |
| except requests.exceptions.HTTPError as ee: | |
| print(f"Warning: Wikipedia API HTTP error {ee}") | |
| return "" | |
| data = r.json() | |
| pages = data.get("query", {}).get("pages", {}) | |
| if not pages: | |
| return "" | |
| text = next(iter(pages.values())).get("extract", "") | |
| return text or "" | |
| def wikipedia_api_query(title: str) -> str: | |
| """Get plain text extract from English Wikipedia for a page title.""" | |
| return _wikipedia_api_query(title) | |
| def wikipedia_search(query: str) -> str: | |
| """Search Wikipedia for encyclopedic knowledge: historical facts, biographies, dates, definitions, figures, scientific information. Provides structured text summaries.""" | |
| try: | |
| wiki = WikipediaAPIWrapper(top_k_results=3, doc_content_chars_max=3000) | |
| return wiki.run(query) | |
| except Exception as e: | |
| return f"Wikipedia error: {e}" | |
| def python_repl(code: str) -> str: | |
| """ | |
| Execute Python code for mathematical calculations, data processing, logic operations, and transformations. | |
| Always use print() to output results. | |
| Examples: print(2**10), print([1,2,3].count(2)), data=[1,2,3]; print(sum(data)/len(data)) | |
| """ | |
| import io, sys | |
| old_stdout = sys.stdout | |
| sys.stdout = io.StringIO() | |
| try: | |
| exec(code, {"math": math, "json": json, "re": re, | |
| "unicodedata": unicodedata, "__builtins__": __builtins__}) | |
| output = sys.stdout.getvalue() | |
| return output.strip() if output.strip() else "Code executed with no output. Use print()." | |
| except Exception as e: | |
| return f"Code error: {e}" | |
| finally: | |
| sys.stdout = old_stdout | |
| def calculator(expression: str) -> str: | |
| """ | |
| Evaluate a mathematical expression quickly. Use for simple arithmetic and compound calculations. | |
| Examples: '2 + 2', '100 * 1.07 ** 5', 'math.sqrt(144)', '(50 + 30) / 2' | |
| """ | |
| try: | |
| return str(eval(expression, {"math": math, "__builtins__": {}})) | |
| except Exception as e: | |
| return f"Calculation error: {e}" | |
| def get_task_file(task_id: str) -> str: | |
| """ | |
| Fetch the file or document attached to a GAIA task by its task_id. | |
| Use this when the question mentions an attached file, document, PDF, or any attachment. | |
| Returns text content for text/JSON files, or indicates binary file type. | |
| """ | |
| try: | |
| import requests as req | |
| url = f"https://agents-course-unit4-scoring.hf.space/files/{task_id}" | |
| response = req.get(url, timeout=15) | |
| if response.status_code == 200: | |
| ct = response.headers.get("Content-Type", "") | |
| if "text" in ct or "json" in ct: | |
| return response.text[:5000] | |
| return f"[Binary file - content-type: {ct}]" | |
| return f"No file found for task {task_id}" | |
| except Exception as e: | |
| return f"Error fetching task file: {e}" | |
| class AgentState(TypedDict): | |
| messages: Annotated[list[AnyMessage], add_messages] | |
| SYSTEM_PROMPT = """You are a highly capable GAIA benchmark solver. Your goal is to answer questions accurately and precisely. | |
| ## How to Solve Questions - Step by Step | |
| 1. **Understand the Question**: Read carefully and identify: | |
| - What type of answer is expected (number, text, list, date, etc.) | |
| - Key constraints or special formats mentioned | |
| - Whether a file or document is attached | |
| 2. **Choose Your Approach**: | |
| - For arithmetic/math: Use `calculator` or `python_repl` | |
| - For current facts/events: Use `web_search` | |
| - For historical/encyclopedic knowledge: Use `wikipedia_search` | |
| - For attached files: Use `get_task_file` | |
| - For complex logic/data processing: Use `python_repl` | |
| 3. **Use Tools Effectively**: | |
| - Search for key facts and verify information from multiple sources | |
| - Extract relevant data from search results | |
| - Perform calculations or transformations | |
| - Cross-check results when possible | |
| 4. **Format Your Final Answer**: | |
| - For numbers: just the number (e.g., "42", "3.14", "-5") | |
| - For text: exact text without extra punctuation (e.g., "Paris", "Monday") | |
| - For lists: comma-separated values (e.g., "item1, item2, item3") | |
| - For dates: use the format specified in the question | |
| - If completely unsure: respond with just "Unknown" | |
| 5. **End Response**: | |
| After your reasoning, output a clean final answer on a new line: | |
| FINAL ANSWER: <your answer> | |
| ## Important Rules | |
| - Never make up facts - always search or calculate | |
| - Verify key numbers and spelling with web search | |
| - If a calculation is involved, always show the work | |
| - Be concise in your reasoning but thorough in verification | |
| """ | |
| def _tool_to_openai_schema(t) -> dict: | |
| """Converte un LangChain tool nel formato tool OpenAI.""" | |
| return { | |
| "type": "function", | |
| "function": { | |
| "name": t.name, | |
| "description": t.description, | |
| "parameters": t.args_schema.schema() if t.args_schema else {"type": "object", "properties": {}}, | |
| } | |
| } | |
| # --- Basic Agent Definition --- | |
| # ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------ | |
| class BasicAgent: | |
| def __init__(self): | |
| print("Initializing agent with HF InferenceClient...") | |
| self.tools_list = [ | |
| web_search, | |
| wikipedia_search, | |
| python_repl, | |
| calculator, | |
| get_task_file, | |
| ] | |
| # Mappa nome → funzione tool per esecuzione | |
| self.tools_by_name = {t.name: t for t in self.tools_list} | |
| hf_token = os.getenv("HF_TOKEN") | |
| if not hf_token: | |
| print("WARNING: HF_TOKEN non impostata. L'agente userà fallback locale e risposte molto limitate.") | |
| self.client = None | |
| else: | |
| # InferenceClient diretto — usa la Serverless Inference API HF | |
| self.client = InferenceClient( | |
| api_key=hf_token, | |
| ) | |
| # Schema OpenAI dei tool per passarli al client | |
| self.tools_schema = [_tool_to_openai_schema(t) for t in self.tools_list] | |
| # Grafo LangGraph per gestire il loop ReAct | |
| builder = StateGraph(AgentState) | |
| builder.add_node("assistant", self._assistant_node) | |
| builder.add_node("tools", ToolNode(self.tools_list)) | |
| builder.add_edge(START, "assistant") | |
| builder.add_conditional_edges("assistant", tools_condition) | |
| builder.add_edge("tools", "assistant") | |
| self.graph = builder.compile() | |
| print("Agent ready.") | |
| def _messages_to_hf_format(self, messages: list) -> list: | |
| """Converte messaggi LangChain nel formato dict che InferenceClient si aspetta.""" | |
| result = [] | |
| for m in messages: | |
| if isinstance(m, SystemMessage): | |
| result.append({"role": "system", "content": m.content}) | |
| elif isinstance(m, HumanMessage): | |
| result.append({"role": "user", "content": m.content}) | |
| elif isinstance(m, AIMessage): | |
| msg = {"role": "assistant", "content": m.content or ""} | |
| # Includi tool_calls se presenti | |
| if m.tool_calls: | |
| msg["tool_calls"] = [ | |
| { | |
| "id": tc["id"], | |
| "type": "function", | |
| "function": { | |
| "name": tc["name"], | |
| "arguments": json.dumps(tc["args"]), | |
| } | |
| } | |
| for tc in m.tool_calls | |
| ] | |
| result.append(msg) | |
| elif isinstance(m, ToolMessage): | |
| result.append({ | |
| "role": "tool", | |
| "tool_call_id": m.tool_call_id, | |
| "content": m.content, | |
| }) | |
| return result | |
| def _assistant_node(self, state: AgentState): | |
| """Nodo assistant: chiama InferenceClient con i tool e restituisce la risposta.""" | |
| sys_msg = SystemMessage(content=SYSTEM_PROMPT) | |
| hf_messages = self._messages_to_hf_format([sys_msg] + state["messages"]) | |
| response = self.client.chat_completion( | |
| model="Qwen/Qwen2.5-72B-Instruct", | |
| messages=hf_messages, | |
| tools=self.tools_schema, | |
| tool_choice="auto", | |
| max_tokens=1000, | |
| temperature=0.1, | |
| ) | |
| choice = response.choices[0].message | |
| # Costruisci AIMessage compatibile con LangGraph | |
| tool_calls = [] | |
| if choice.tool_calls: | |
| for tc in choice.tool_calls: | |
| tool_calls.append({ | |
| "id": tc.id, | |
| "name": tc.function.name, | |
| "args": json.loads(tc.function.arguments), | |
| "type": "tool_call", | |
| }) | |
| ai_message = AIMessage( | |
| content=choice.content or "", | |
| tool_calls=tool_calls, | |
| ) | |
| return {"messages": [ai_message]} | |
| def _local_fallback_answer(self, question: str) -> str: | |
| """ | |
| Minimal fallback when inference client is unavailable. | |
| Attempts basic arithmetic only, otherwise returns Unknown. | |
| """ | |
| q = question.lower().strip() | |
| # Try simple arithmetic if it looks like a math problem | |
| if re.search(r"(?:how\s+many|calculate|compute|what\s+is).*\d+", q): | |
| try: | |
| # Try to extract and evaluate a simple expression | |
| numbers = re.findall(r"\d+\.?\d*", question) | |
| if len(numbers) >= 2: | |
| # Don't try to hardcode logic - just return Unknown | |
| pass | |
| except Exception: | |
| pass | |
| return "Unknown" | |
| def __call__(self, question: str) -> str: | |
| print(f"Agent received question (first 100 chars): {question[:100]}...") | |
| if self.client is None: | |
| print("No HF InferenceClient configured; using local fallback logic") | |
| return self._local_fallback_answer(question) | |
| q_lower = question.lower().strip() | |
| # Numeric math shortcut: explicit arithmetic detection reduces hallucination. | |
| arithmetic = self._extract_arithmetic_expression(question) | |
| if arithmetic: | |
| calc = calculator(arithmetic) | |
| if not calc.startswith("Calculation error"): | |
| normalized = self._normalize_answer(calc) | |
| print(f"Arithmetic shortcut using calculator: {arithmetic} -> {normalized}") | |
| return normalized | |
| # The main RL loop | |
| answer = self._run_agent(question) | |
| # Retry with explicit “Unknown” handling and chain-of-thought guidance | |
| if answer == "Unknown": | |
| print("Got Unknown on first pass; retrying with more explicit reasoning request") | |
| replay_question = question + "\n\nPlease reason step by step with tool calls and provide FINAL ANSWER only." # gentle prompt nudge | |
| answer = self._run_agent(replay_question) | |
| print(f"Agent returning answer: '{answer}'") | |
| return answer | |
| def _run_agent(self, question: str) -> str: | |
| try: | |
| result = self.graph.invoke({"messages": [HumanMessage(content=question)]}) | |
| last_message = result["messages"][-1] | |
| response_text = last_message.content if isinstance(last_message, AIMessage) else str(last_message) | |
| print(f"Agent raw output (first 300 chars): {response_text[:300]}...") | |
| return self._extract_answer(response_text) | |
| except Exception as e: | |
| error_message = str(e) | |
| print(f"Agent error during run: {error_message}") | |
| if "402" in error_message or "Payment Required" in error_message: | |
| fallback = self._local_fallback_answer(question) | |
| print(f"Payment required detected; using local fallback answer: {fallback}") | |
| return fallback | |
| return "Unknown" | |
| def _extract_answer(self, text: str) -> str: | |
| """ | |
| Extract the final answer from agent output using multiple strategies. | |
| """ | |
| # Strategy 1: Look for explicit "FINAL ANSWER:" marker | |
| match = re.search(r"FINAL ANSWER:\s*(.+?)(?:\n|$)", text, re.IGNORECASE) | |
| if match: | |
| answer = self._normalize_answer(match.group(1).strip()) | |
| if answer and answer != "Unknown": | |
| return answer | |
| # Strategy 2: Look at the last few lines | |
| lines = [line.strip() for line in text.split('\n') if line.strip()] | |
| if lines: | |
| for candidate in reversed(lines[-4:]): | |
| if candidate and not any(phrase in candidate.lower() for phrase in ["i'm not sure", "error", "failed", "final answer"]): | |
| normalized = self._normalize_answer(candidate) | |
| if normalized and normalized != "Unknown": | |
| return normalized | |
| # Fallback | |
| return "Unknown" | |
| def _normalize_answer(self, answer: str) -> str: | |
| """Normalize answer text (strip punctuation, normalize choices etc.).""" | |
| answer_clean = answer.strip().strip('"\'').rstrip('.?,;') | |
| if not answer_clean: | |
| return "Unknown" | |
| # Multiple-choice token: take plain option text if present | |
| mc = re.match(r"^([A-D])\s*[:\)]\s*(.+)$", answer_clean, re.IGNORECASE) | |
| if mc: | |
| return mc.group(2).strip() | |
| # Numeric decision: enforce numeric format for numeric questions | |
| if re.match(r"^-?\d+(\.\d+)?$", answer_clean): | |
| return answer_clean | |
| return answer_clean | |
| def _extract_arithmetic_expression(self, question: str) -> Optional[str]: | |
| """Extract simple arithmetic expression candidate from a question for calculator use.""" | |
| m = re.search(r"([-+]?\d+(?:\.\d+)?(?:\s*[-+*/]\s*\d+(?:\.\d+)?)+)", question) | |
| if not m: | |
| return None | |
| expr = m.group(1).replace("^", "**") | |
| if re.search(r"[a-zA-Z]", expr): | |
| return None | |
| return expr | |
| def _detect_question_type(self, question: str) -> str: | |
| q = question.lower().strip() | |
| if any(tok in q for tok in ["how many", "calculate", "compute", "sum", "difference", "times", "per cent", "%"]): | |
| return "numeric" | |
| if any(tok in q for tok in ["when", "year", "date", "born", "died"]): | |
| return "date" | |
| if any(tok in q for tok in ["which of the following", "option", "choose", "select"]): | |
| return "multiple_choice" | |
| return "factual" | |
| 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 | |
| def build_gradio_ui(): | |
| 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) | |
| results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True) | |
| run_button.click( | |
| fn=run_and_submit_all, | |
| outputs=[status_output, results_table] | |
| ) | |
| return demo | |
| 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") | |
| 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(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 = build_gradio_ui() | |
| demo.launch(debug=True, share=False) | |