commit

app.py

@@ -1,155 +1,349 @@
 import os
-import re
 import time
+import random
 import gradio as gr
 import requests
+import inspect
 import pandas as pd
-from typing import TypedDict
-from langgraph.graph import StateGraph, START, END
+from typing import Optional
+
 from langchain_openai import ChatOpenAI
 from langchain_core.messages import SystemMessage, HumanMessage
 
-# -------------------------
-# Constants
-# -------------------------
+# (Keep Constants as is)
+# --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# -------------------------
-# Utils
-# -------------------------
+# =========================================================
+# Answer cleaning (EXACT MATCH 최적화)
+# =========================================================
 def clean_answer(text: str) -> str:
     if not text:
         return ""
+
     s = text.strip()
-    s = re.sub(r"^(Final answer:|Answer:)\s*", "", s, flags=re.I)
+
+    # 흔한 접두/포맷 제거
+    s = re.sub(r"^(final\s*answer|answer)\s*:\s*", "", s, flags=re.IGNORECASE).strip()
+
+    # 코드블록/마크다운 제거
+    s = s.strip("`").strip()
+
+    # 여러 줄이면 첫 줄만
     s = s.splitlines()[0].strip()
-    s = s.strip("\"'`")
-    if s.endswith(".") and len(s) > 1:
-        s = s[:-1]
+
+    # 양끝 따옴표 제거
+    s = s.strip("\"'")
+
+    # 끝에 . 하나 붙는 버릇 제거 (단, 약어/소수점은 건드리면 위험하니 매우 보수적으로)
+    if len(s) > 1 and s.endswith(".") and not re.search(r"\d\.$", s):
+        s = s[:-1].strip()
+
+    # 불필요한 공백 정리
+    s = re.sub(r"\s+", " ", s).strip()
+
     return s
 
-# -------------------------
-# State
-# -------------------------
-class AgentState(TypedDict):
-    question: str
-    answer: str
+# =========================================================
+# Robust request wrapper for GAIA server (429 대응)
+# =========================================================
+def get_with_backoff(url: str, timeout: int = 15, max_retries: int = 6) -> requests.Response:
+    for i in range(max_retries):
+        try:
+            r = requests.get(url, timeout=timeout)
+            if r.status_code == 429:
+                # 지수 백오프 + 지터
+                sleep_s = min(30, (2 ** i) + random.uniform(0, 1.5))
+                print(f"[WARN] 429 Too Many Requests. Sleeping {sleep_s:.2f}s then retry {i+1}/{max_retries}...")
+                time.sleep(sleep_s)
+                continue
+            r.raise_for_status()
+            return r
+        except requests.exceptions.RequestException as e:
+            if i == max_retries - 1:
+                raise
+            sleep_s = min(20, (2 ** i) + random.uniform(0, 1.0))
+            print(f"[WARN] GET failed: {e}. Sleeping {sleep_s:.2f}s then retry {i+1}/{max_retries}...")
+            time.sleep(sleep_s)
+    raise RuntimeError("get_with_backoff exhausted retries")
+
+def post_with_backoff(url: str, json_data: dict, timeout: int = 60, max_retries: int = 5) -> requests.Response:
+    for i in range(max_retries):
+        try:
+            r = requests.post(url, json=json_data, timeout=timeout)
+            if r.status_code == 429:
+                sleep_s = min(30, (2 ** i) + random.uniform(0, 1.5))
+                print(f"[WARN] 429 Too Many Requests (POST). Sleeping {sleep_s:.2f}s then retry {i+1}/{max_retries}...")
+                time.sleep(sleep_s)
+                continue
+            r.raise_for_status()
+            return r
+        except requests.exceptions.RequestException as e:
+            if i == max_retries - 1:
+                raise
+            sleep_s = min(20, (2 ** i) + random.uniform(0, 1.0))
+            print(f"[WARN] POST failed: {e}. Sleeping {sleep_s:.2f}s then retry {i+1}/{max_retries}...")
+            time.sleep(sleep_s)
+    raise RuntimeError("post_with_backoff exhausted retries")
+
+# =========================================================
+# LLM setup (OpenAI)
+# =========================================================
+# Space Secrets에 OPENAI_API_KEY 필요
+OPENAI_API_KEY = os.getenv("OPENAI_API_KEY", "")
+if OPENAI_API_KEY:
+    os.environ["OPENAI_API_KEY"] = OPENAI_API_KEY
 
-# -------------------------
-# LLM
-# -------------------------
 llm = ChatOpenAI(
     model="gpt-4o-mini",
     temperature=0,
-    max_tokens=128
+    max_tokens=96,
 )
 
-SYSTEM_PROMPT = """
-You are solving GAIA benchmark questions.
+SYSTEM_PROMPT = """You are solving GAIA benchmark questions.
 
-Rules (VERY IMPORTANT):
-- You MUST reason step by step internally.
-- NEVER show your reasoning.
-- Output ONLY the final answer.
-- No explanation.
-- No formatting.
-- No extra words.
+Hard rules:
+- Think step by step internally, but DO NOT reveal your reasoning.
+- Verify arithmetic, units, dates, and entity names before finalizing.
+- Output ONLY the final answer (exactly what should be matched).
+- No explanation. No prefixes. No punctuation unless required by the answer itself.
+- If the answer is a number/date/name, output it in the simplest canonical form.
 """
 
-# -------------------------
-# Node
-# -------------------------
-def solve_question(state: AgentState) -> dict:
-    question = state["question"]
+def build_user_prompt(question: str) -> str:
+    return f"""Question:
+{question}
 
-    messages = [
+Return ONLY the final answer.
+"""
+
+# =========================================================
+# 2-pass solve: (1) answer (2) self-check and possibly revise
+# =========================================================
+def solve_with_selfcheck(question: str) -> str:
+    # Pass 1: initial solve
+    msg1 = [
         SystemMessage(content=SYSTEM_PROMPT),
-        HumanMessage(content=question)
+        HumanMessage(content=build_user_prompt(question)),
     ]
+    r1 = llm.invoke(msg1)
+    a1 = clean_answer(getattr(r1, "content", "") or "")
+
+    # Pass 2: self-check (짧게 검증만)
+    # - GAIA는 "정답만"을 요구하므로, 검증도 출력은 정답만 하게 강제
+    check_prompt = f"""You previously answered: {a1}
 
-    resp = llm.invoke(messages)
-    answer = clean_answer(resp.content)
+Now do a silent verification. If the answer is wrong or not in canonical exact-match form, output the corrected final answer.
+If it is correct, output exactly the same answer again.
 
-    return {"answer": answer}
+Question:
+{question}
 
-# -------------------------
-# Graph
-# -------------------------
-builder = StateGraph(AgentState)
-builder.add_node("solve", solve_question)
-builder.add_edge(START, "solve")
-builder.add_edge("solve", END)
-agent_graph = builder.compile()
+Return ONLY the final answer.
+"""
+    msg2 = [
+        SystemMessage(content=SYSTEM_PROMPT),
+        HumanMessage(content=check_prompt),
+    ]
+    r2 = llm.invoke(msg2)
+    a2 = clean_answer(getattr(r2, "content", "") or "")
 
-# -------------------------
-# Wrapper
-# -------------------------
+    # 둘 다 비었으면 실패 처리
+    if not a2 and a1:
+        return a1
+    return a2
+
+# =========================================================
+# Basic Agent Definition (템플릿 유지, 여기만 “진짜”로 바꿈)
+# =========================================================
 class BasicAgent:
+    def __init__(self):
+        print("BasicAgent initialized (LLM + self-check).")
+
     def __call__(self, question: str) -> str:
-        result = agent_graph.invoke({"question": question, "answer": ""})
-        return result["answer"]
+        print(f"Agent received question (first 50 chars): {question[:50]}...")
 
-# -------------------------
-# Runner
-# -------------------------
-def run_and_submit_all(profile: gr.OAuthProfile | None):
-    if not profile:
-        return "Please login first.", None
+        try:
+            answer = solve_with_selfcheck(question)
+        except Exception as e:
+            # LLM 에러가 나면 빈 답 내면 0점이니, 최소한 에러를 로깅하고 빈 문자열 반환
+            # (여기서 다른 fallback 넣고 싶으면 넣을 수 있음)
+            print(f"[ERROR] LLM call failed: {e}")
+            answer = ""
 
-    username = profile.username
-    questions_url = f"{DEFAULT_API_URL}/questions"
-    submit_url = f"{DEFAULT_API_URL}/submit"
+        print(f"Agent returning answer: {answer}")
+        return answer
 
-    agent = BasicAgent()
+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
 
-    # fetch once
-    questions = requests.get(questions_url).json()
+    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
 
-    answers_payload = []
-    log = []
+    api_url = DEFAULT_API_URL
+    questions_url = f"{api_url}/questions"
+    submit_url = f"{api_url}/submit"
 
-    for q in questions:
+    # 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:
-            ans = agent(q["question"])
-            answers_payload.append({
-                "task_id": q["task_id"],
-                "submitted_answer": ans
-            })
-            log.append({
-                "task_id": q["task_id"],
-                "answer": ans
-            })
+            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:
-            log.append({
-                "task_id": q["task_id"],
-                "answer": f"ERROR: {e}"
-            })
-
-    submission = {
-        "username": username,
-        "agent_code": "https://huggingface.co/spaces/ahnhs2k/Agents_Final_Assignment_",
-        "answers": answers_payload
-    }
-
-    res = requests.post(submit_url, json=submission).json()
-
-    status = (
-        f"Score: {res.get('score')}% "
-        f"({res.get('correct_count')}/{res.get('total_attempted')})\n"
-        f"{res.get('message')}"
-    )
+             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
 
-    return status, pd.DataFrame(log)
 
-# -------------------------
-# UI
-# -------------------------
+# --- Build Gradio Interface using Blocks ---
 with gr.Blocks() as demo:
-    gr.Markdown("# GAIA Level-1 Agent")
+    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()
-    btn = gr.Button("Run & Submit")
-    out = gr.Textbox(lines=4)
-    table = gr.DataFrame()
-    btn.click(run_and_submit_all, outputs=[out, table])
 
-demo.launch()
+    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)