Skip to content

  • 体验新版
    • 正在加载...
  • 登录
  • PaddlePaddle
  • PaddleDetection
  • Issue
  • #858

P
PaddleDetection
  • 项目概览

PaddlePaddle / PaddleDetection
大约 2 年 前同步成功

通知 708
Star 11112
Fork 2696
  • 代码
    • 文件
    • 提交
    • 分支
    • Tags
    • 贡献者
    • 分支图
    • Diff
  • Issue 184
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 40
  • Wiki 0
    • Wiki
  • 分析
    • 仓库
    • DevOps
  • 项目成员
  • Pages
P
PaddleDetection
  • 项目概览
    • 项目概览
    • 详情
    • 发布
  • 仓库
    • 仓库
    • 文件
    • 提交
    • 分支
    • 标签
    • 贡献者
    • 分支图
    • 比较
  • Issue 184
    • Issue 184
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 40
    • 合并请求 40
  • Pages
  • 分析
    • 分析
    • 仓库分析
    • DevOps
  • Wiki 0
    • Wiki
  • 成员
    • 成员
  • 收起侧边栏
  • 动态
  • 分支图
  • 创建新Issue
  • 提交
  • Issue看板
已关闭
开放中
Opened 6月 03, 2020 by saxon_zh@saxon_zhGuest

多线程处理时,多卡执行tools/eval.py报错

Created by: flishwang

EvalReader中use_process设为false,且batch_size设为1时,且gpu数量大于1时,程序在退出时报错。 use_process设为true,或者batch_size设为2,或data_parrallel部分只设置1个CUDAPlace时, 程序都能正常运行。

报错信息如下:

2020-06-03 11:11:54,109-INFO: start loading proposals 2020-06-03 11:11:54,229-INFO: loading roidb 2012_test 2020-06-03 11:11:58,729-INFO: load roidb from scope 2012_test 2020-06-03 11:11:58,729-INFO: finish loading roidbs, total num = 293 2020-06-03 11:11:58,730-INFO: set max batches to 0 2020-06-03 11:11:58,731-INFO: places would be ommited when DataLoader is not iterable W0603 11:11:58.798434 13498 device_context.cc:252] Please NOTE: device: 0, CUDA Capability: 70, Driver API Version: 10.1, Runtime API Version: 10.0 W0603 11:11:59.370779 13498 device_context.cc:260] device: 0, cuDNN Version: 7.6. 2020-06-03 11:12:35,821-INFO: Test iter 0 2020-06-03 11:12:51,766-INFO: Test iter 100 2020-06-03 11:13:06,355-INFO: Test iter 200 2020-06-03 11:13:19,731-INFO: Test finish iter 293 2020-06-03 11:13:19,731-INFO: Total iteration: 293, inference time: 5.194278110324853 batch/s. 2020-06-03 11:13:19,791-INFO: Start evaluate... 2020-06-03 11:13:19,872-INFO: Accumulating evaluatation results... 2020-06-03 11:13:19,875-INFO: mAP@0.20(integral) = 93.30 2020-06-03 11:13:19,875-INFO: Start evaluate... 2020-06-03 11:13:19,956-INFO: Accumulating evaluatation results... 2020-06-03 11:13:19,963-INFO: mAP@0.50(integral) = 90.65 2020-06-03 11:13:19,963-INFO: Start evaluate... 2020-06-03 11:13:20,064-INFO: Accumulating evaluatation results... 2020-06-03 11:13:20,076-INFO: mAP@0.70(integral) = 71.76 terminate called without an active exception W0603 11:13:20.411875 13642 init.cc:216] Warning: PaddlePaddle catches a failure signal, it may not work properly W0603 11:13:20.411936 13642 init.cc:218] You could check whether you killed PaddlePaddle thread/process accidentally or report the case to PaddlePaddle W0603 11:13:20.411949 13642 init.cc:221] The detail failure signal is:

W0603 11:13:20.411965 13642 init.cc:224] *** Aborted at 1591154000 (unix time) try "date -d @1591154000" if you are using GNU date *** W0603 11:13:20.416925 13642 init.cc:224] PC: @ 0x0 (unknown) W0603 11:13:20.417045 13642 init.cc:224] *** SIGABRT (@0x2716000034ba) received by PID 13498 (TID 0x7f7e69feb700) from PID 13498; stack trace: *** W0603 11:13:20.421667 13642 init.cc:224] @ 0x7f8124ef5390 (unknown) W0603 11:13:20.426170 13642 init.cc:224] @ 0x7f8124b4f428 gsignal W0603 11:13:20.429155 13642 init.cc:224] @ 0x7f8124b5102a abort W0603 11:13:20.455327 13642 init.cc:224] @ 0x7f807560184a __gnu_cxx::__verbose_terminate_handler() W0603 11:13:20.457710 13642 init.cc:224] @ 0x7f80755fff47 __cxxabiv1::__terminate() W0603 11:13:20.460386 13642 init.cc:224] @ 0x7f80755fff7d std::terminate() W0603 11:13:20.462821 13642 init.cc:224] @ 0x7f80755ffc5a __gxx_personality_v0 W0603 11:13:20.466262 13642 init.cc:224] @ 0x7f81220c1b97 _Unwind_ForcedUnwind_Phase2 W0603 11:13:20.468724 13642 init.cc:224] @ 0x7f81220c1e7d _Unwind_ForcedUnwind W0603 11:13:20.471596 13642 init.cc:224] @ 0x7f8124ef4070 __GI___pthread_unwind W0603 11:13:20.474467 13642 init.cc:224] @ 0x7f8124eec845 __pthread_exit W0603 11:13:20.533114 13642 init.cc:224] @ 0x55ad710df059 PyThread_exit_thread W0603 11:13:20.533632 13642 init.cc:224] @ 0x55ad70f64c10 PyEval_RestoreThread.cold.799 W0603 11:13:20.537024 13642 init.cc:224] @ 0x7f8111691cde (unknown) W0603 11:13:20.538024 13642 init.cc:224] @ 0x55ad71065ab4 _PyMethodDef_RawFastCallKeywords W0603 11:13:20.538980 13642 init.cc:224] @ 0x55ad71065bd1 _PyCFunction_FastCallKeywords W0603 11:13:20.539932 13642 init.cc:224] @ 0x55ad710cc57b _PyEval_EvalFrameDefault W0603 11:13:20.540813 13642 init.cc:224] @ 0x55ad71011389 _PyEval_EvalCodeWithName W0603 11:13:20.541718 13642 init.cc:224] @ 0x55ad710124c5 _PyFunction_FastCallDict W0603 11:13:20.542596 13642 init.cc:224] @ 0x55ad71031a73 _PyObject_Call_Prepend W0603 11:13:20.543094 13642 init.cc:224] @ 0x55ad7107927a slot_tp_call W0603 11:13:20.543982 13642 init.cc:224] @ 0x55ad7107a2db _PyObject_FastCallKeywords W0603 11:13:20.544926 13642 init.cc:224] @ 0x55ad710cc146 _PyEval_EvalFrameDefault W0603 11:13:20.545817 13642 init.cc:224] @ 0x55ad710123fb _PyFunction_FastCallDict W0603 11:13:20.546566 13642 init.cc:224] @ 0x55ad71031a73 _PyObject_Call_Prepend W0603 11:13:20.546892 13642 init.cc:224] @ 0x55ad7107927a slot_tp_call W0603 11:13:20.547472 13642 init.cc:224] @ 0x55ad7107a2db _PyObject_FastCallKeywords W0603 11:13:20.548089 13642 init.cc:224] @ 0x55ad710cca39 _PyEval_EvalFrameDefault W0603 11:13:20.548661 13642 init.cc:224] @ 0x55ad71011389 _PyEval_EvalCodeWithName W0603 11:13:20.549237 13642 init.cc:224] @ 0x55ad710124c5 _PyFunction_FastCallDict W0603 11:13:20.549818 13642 init.cc:224] @ 0x55ad71031a73 _PyObject_Call_Prepend W0603 11:13:20.550443 13642 init.cc:224] @ 0x55ad71023fde PyObject_Call Aborted (core dumped)

指派人
分配到
无
里程碑
无
分配里程碑
工时统计
无
截止日期
无
标识: paddlepaddle/PaddleDetection#858
渝ICP备2023009037号

京公网安备11010502055752号

网络110报警服务 Powered by GitLab CE v13.7
开源知识
Git 入门 Pro Git 电子书 在线学 Git
Markdown 基础入门 IT 技术知识开源图谱
帮助
使用手册 反馈建议 博客
《GitCode 隐私声明》 《GitCode 服务条款》 关于GitCode
Powered by GitLab CE v13.7