import os import gradio as gr import pandas as pd import requests from agent import GaiaAgent from config import ( DEFAULT_API_URL, HF_PLANNER_USE_RESPONSE_FORMAT, HF_TEXT_MODEL, HF_VISION_MODEL, ) def run_and_submit_all(profile: gr.OAuthProfile | None): """ 拉取全部问题，使用 GaiaAgent 逐题作答，提交全部答案，并展示评测结果。 """ space_id = os.getenv("SPACE_ID") if profile: username = f"{profile.username}" print(f"用户已登录：{username}") else: print("用户未登录。") return "请先点击按钮登录 Hugging Face。", None api_url = DEFAULT_API_URL questions_url = f"{api_url}/questions" submit_url = f"{api_url}/submit" try: agent = GaiaAgent() except Exception as e: print(f"实例化 Agent 时出错：{e}") return f"Agent 初始化失败：{e}", None agent_code = ( f"https://huggingface.co/spaces/{space_id}/tree/main" if space_id else "本地运行，未提供 SPACE_ID。" ) print(agent_code) print(f"正在从以下地址拉取问题：{questions_url}") try: response = requests.get(questions_url, timeout=15) response.raise_for_status() questions_data = response.json() if not questions_data: print("拉取到的问题列表为空。") return "拉取到的问题列表为空，或返回格式无效。", None print(f"已拉取 {len(questions_data)} 个问题。") except requests.exceptions.RequestException as e: print(f"拉取问题时发生网络错误：{e}") return f"拉取问题失败：{e}", None except requests.exceptions.JSONDecodeError as e: print(f"解析问题接口的 JSON 响应时出错：{e}") print(f"响应文本：{response.text[:500]}") return f"解析问题接口响应失败：{e}", None except Exception as e: print(f"拉取问题时发生未预期错误：{e}") return f"拉取问题时发生未预期错误：{e}", None results_log = [] answers_payload = [] print(f"正在让 Agent 处理 {len(questions_data)} 个问题...") for item in questions_data: task_id = item.get("task_id") question_text = item.get("question") file_name = item.get("file_name") or "" if not task_id or question_text is None: print(f"跳过缺少 task_id 或 question 的条目：{item}") continue try: result = agent.answer_task(question_text, task_id=task_id, file_name=file_name) submitted_answer = result.answer or "无法确定" answers_payload.append( {"task_id": task_id, "submitted_answer": submitted_answer} ) results_log.append( { "任务 ID": task_id, "附件": file_name, "问题": question_text, "提交答案": submitted_answer, "来源": result.source, "置信度": result.confidence, "错误": result.error, } ) except Exception as e: print(f"Agent 处理任务 {task_id} 时出错：{e}") results_log.append( { "任务 ID": task_id, "附件": file_name, "问题": question_text, "提交答案": f"AGENT ERROR: {e}", "来源": "agent.exception", "置信度": "low", "错误": str(e), } ) if not answers_payload: print("Agent 没有生成任何可提交的答案。") return "Agent 没有生成任何可提交的答案。", pd.DataFrame(results_log) submission_data = { "username": username.strip(), "agent_code": agent_code, "answers": answers_payload, } status_update = f"Agent 已完成作答。正在为用户 '{username}' 提交 {len(answers_payload)} 个答案..." print(status_update) print(f"正在向以下地址提交 {len(answers_payload)} 个答案：{submit_url}") try: response = requests.post(submit_url, json=submission_data, timeout=60) response.raise_for_status() result_data = response.json() final_status = ( f"提交成功！\n" f"用户：{result_data.get('username')}\n" f"总分：{result_data.get('score', 'N/A')}% " f"（{result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} 正确）\n" f"消息：{result_data.get('message', '未收到消息。')}" ) print("提交成功。") return final_status, pd.DataFrame(results_log) except requests.exceptions.HTTPError as e: error_detail = f"服务器返回状态码 {e.response.status_code}。" try: error_json = e.response.json() error_detail += f" 详情：{error_json.get('detail', e.response.text)}" except requests.exceptions.JSONDecodeError: error_detail += f" 响应：{e.response.text[:500]}" status_message = f"提交失败：{error_detail}" print(status_message) return status_message, pd.DataFrame(results_log) except requests.exceptions.Timeout: status_message = "提交失败：请求超时。" print(status_message) return status_message, pd.DataFrame(results_log) except requests.exceptions.RequestException as e: status_message = f"提交失败：网络错误 - {e}" print(status_message) return status_message, pd.DataFrame(results_log) except Exception as e: status_message = f"提交过程中发生未预期错误：{e}" print(status_message) return status_message, pd.DataFrame(results_log) with gr.Blocks() as demo: gr.Markdown("# LangGraph GAIA Agent 评测运行器") gr.Markdown( f""" **当前策略：** 1. LangGraph 先调用 LLM 做题型分类，明确应该使用哪类工具。 2. 工作流按题型进入规则、Python、Excel、Wikipedia、体育统计、网页读取/搜索等处理节点。 3. 工具结果统一交给 LLM 做最终答案格式化；LLM 失败时使用工具候选答案兜底。 4. 未识别类型或工具无结果时进入兜底节点，按低成本工具顺序尝试。 5. 音频、视频和棋局图工具当前禁用；这类题会返回“无法确定”。 6. 当前文本模型：`{HF_TEXT_MODEL}`；当前视觉模型：`{HF_VISION_MODEL or "未启用"}`。 7. 强制 JSON response_format：`{HF_PLANNER_USE_RESPONSE_FORMAT}`。 **提交前检查：** Space Secrets 至少需要 `HF_TOKEN`，且该 token 必须允许调用 Inference Providers。 """ ) gr.LoginButton() run_button = gr.Button("运行评测并提交全部答案") status_output = gr.Textbox(label="运行状态 / 提交结果", lines=5, interactive=False) results_table = gr.DataFrame(label="问题与 Agent 答案", wrap=True) run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table]) if __name__ == "__main__": print("\n" + "-" * 30 + " 应用启动中 " + "-" * 30) space_host_startup = os.getenv("SPACE_HOST") space_id_startup = os.getenv("SPACE_ID") if space_host_startup: print(f"找到 SPACE_HOST：{space_host_startup}") print(f" 运行地址应为：https://{space_host_startup}.hf.space") else: print("未找到 SPACE_HOST 环境变量（可能是在本地运行）。") if space_id_startup: print(f"找到 SPACE_ID：{space_id_startup}") print(f" 仓库地址：https://huggingface.co/spaces/{space_id_startup}") print(f" 代码树地址：https://huggingface.co/spaces/{space_id_startup}/tree/main") else: print("未找到 SPACE_ID 环境变量（可能是在本地运行）。无法确定仓库地址。") print("-" * (60 + len(" 应用启动中 ")) + "\n") print("正在启动 LangGraph GAIA Agent 评测 Gradio 界面...") demo.launch(debug=True, share=False)