models icon indicating copy to clipboard operation
models copied to clipboard

Published 20 hours ago • verified publisher icon PaddlePaddle

运行tools/eval.py报错

Open COST-97 opened this issue 4 years ago • 2 comments

模型成功使用tools/train.py训练， 但是运行tools/eval.py报错。 运行代码：

‘%cd ~
%cd PaddleDetection/
!export CUDA_VISIBLE_DEVICES=0
!python -u tools/eval.py -c configs/yolov3_darknet_voc.yml \
                    -o weights=output/yolov3_darknet_voc/model_final/model_final \’

出错信息如下：

/home/aistudio
/home/aistudio/PaddleDetection
2020-03-17 19:17:51,257-INFO: 555 samples in file dataset/underwater_data/val.txt
2020-03-17 19:17:51,258-INFO: places would be ommited when DataLoader is not iterable
W0317 19:17:52.077756   559 device_context.cc:237] Please NOTE: device: 0, CUDA Capability: 70, Driver API Version: 9.2, Runtime API Version: 9.0
W0317 19:17:52.081887   559 device_context.cc:245] device: 0, cuDNN Version: 7.3.
2020-03-17 19:17:53,552-INFO: Loading parameters from output/yolov3_darknet_voc/model_final/model_final...
I0317 19:17:56.293087   559 parallel_executor.cc:440] The Program will be executed on CUDA using ParallelExecutor, 1 cards are used, so 1 programs are executed in parallel.
I0317 19:17:56.304033   559 build_strategy.cc:365] SeqOnlyAllReduceOps:0, num_trainers:1
I0317 19:17:56.312638   559 parallel_executor.cc:307] Inplace strategy is enabled, when build_strategy.enable_inplace = True
I0317 19:17:56.318989   559 parallel_executor.cc:375] Garbage collection strategy is enabled, when FLAGS_eager_delete_tensor_gb = 0
2020-03-17 19:17:56,902-INFO: Test iter 0
2020-03-17 19:18:03,758-INFO: Test finish iter 69
2020-03-17 19:18:03,759-INFO: Total number of images: 546, inference time: 73.04237257565066 fps.
2020-03-17 19:18:03,759-INFO: Start evaluate...
2020-03-17 19:18:04,487-INFO: Accumulating evaluatation results...
2020-03-17 19:18:04,535-INFO: mAP(0.50, 11point) = 24.81
terminate called without an active exception
W0317 19:18:04.659145   628 init.cc:209] Warning: PaddlePaddle catches a failure signal, it may not work properly
W0317 19:18:04.659191   628 init.cc:211] You could check whether you killed PaddlePaddle thread/process accidentally or report the case to PaddlePaddle
W0317 19:18:04.659196   628 init.cc:214] The detail failure signal is:

W0317 19:18:04.659202   628 init.cc:217] *** Aborted at 1584443884 (unix time) try "date -d @1584443884" if you are using GNU date ***
W0317 19:18:04.661578   628 init.cc:217] PC: @                0x0 (unknown)
W0317 19:18:04.661710   628 init.cc:217] *** SIGABRT (@0x3e80000022f) received by PID 559 (TID 0x7f7e2d7fe700) from PID 559; stack trace: ***
W0317 19:18:04.663616   628 init.cc:217]     @     0x7f7f12139390 (unknown)
W0317 19:18:04.665478   628 init.cc:217]     @     0x7f7f11d93428 gsignal
W0317 19:18:04.667274   628 init.cc:217]     @     0x7f7f11d9502a abort
W0317 19:18:04.668453   628 init.cc:217]     @     0x7f7ed2d0084a __gnu_cxx::__verbose_terminate_handler()
W0317 19:18:04.669427   628 init.cc:217]     @     0x7f7ed2cfef47 __cxxabiv1::__terminate()
W0317 19:18:04.670521   628 init.cc:217]     @     0x7f7ed2cfef7d std::terminate()
W0317 19:18:04.671491   628 init.cc:217]     @     0x7f7ed2cfec5a __gxx_personality_v0
W0317 19:18:04.672348   628 init.cc:217]     @     0x7f7ed2ff1b97 _Unwind_ForcedUnwind_Phase2
W0317 19:18:04.673223   628 init.cc:217]     @     0x7f7ed2ff1e7d _Unwind_ForcedUnwind
W0317 19:18:04.674988   628 init.cc:217]     @     0x7f7f12138070 __GI___pthread_unwind
W0317 19:18:04.676718   628 init.cc:217]     @     0x7f7f12130845 __pthread_exit
W0317 19:18:04.677220   628 init.cc:217]     @     0x55b12b95be59 PyThread_exit_thread
W0317 19:18:04.677367   628 init.cc:217]     @     0x55b12b7e1c17 PyEval_RestoreThread.cold.798
W0317 19:18:04.677793   628 init.cc:217]     @     0x7f7ec7b8008d (unknown)
W0317 19:18:04.678279   628 init.cc:217]     @     0x55b12b8dd7e6 _PyMethodDef_RawFastCallKeywords
W0317 19:18:04.678741   628 init.cc:217]     @     0x55b12b8dd861 _PyCFunction_FastCallKeywords
W0317 19:18:04.679210   628 init.cc:217]     @     0x55b12b9497cc _PyEval_EvalFrameDefault
W0317 19:18:04.679639   628 init.cc:217]     @     0x55b12b88d539 _PyEval_EvalCodeWithName
W0317 19:18:04.680107   628 init.cc:217]     @     0x55b12b88e635 _PyFunction_FastCallDict
W0317 19:18:04.680549   628 init.cc:217]     @     0x55b12b8ace53 _PyObject_Call_Prepend
W0317 19:18:04.680773   628 init.cc:217]     @     0x55b12b8e4a3a slot_tp_call
W0317 19:18:04.681227   628 init.cc:217]     @     0x55b12b8e58fb _PyObject_FastCallKeywords
W0317 19:18:04.681691   628 init.cc:217]     @     0x55b12b948e86 _PyEval_EvalFrameDefault
W0317 19:18:04.682130   628 init.cc:217]     @     0x55b12b88e56b _PyFunction_FastCallDict
W0317 19:18:04.682551   628 init.cc:217]     @     0x55b12b8ace53 _PyObject_Call_Prepend
W0317 19:18:04.682771   628 init.cc:217]     @     0x55b12b8e4a3a slot_tp_call
W0317 19:18:04.683207   628 init.cc:217]     @     0x55b12b8e58fb _PyObject_FastCallKeywords
W0317 19:18:04.683665   628 init.cc:217]     @     0x55b12b9496e8 _PyEval_EvalFrameDefault
W0317 19:18:04.684094   628 init.cc:217]     @     0x55b12b88d539 _PyEval_EvalCodeWithName
W0317 19:18:04.684522   628 init.cc:217]     @     0x55b12b88e635 _PyFunction_FastCallDict
W0317 19:18:04.684940   628 init.cc:217]     @     0x55b12b8ace53 _PyObject_Call_Prepend
W0317 19:18:04.685431   628 init.cc:217]     @     0x55b12b89fdbe PyObject_Call
Aborted (core dumped)

COST-97 avatar Mar 17 '20 11:03 COST-97 

2020-03-17 19:17:56,902-INFO: Test iter 0
2020-03-17 19:18:03,758-INFO: Test finish iter 69
2020-03-17 19:18:03,759-INFO: Total number of images: 546, inference time: 73.04237257565066 fps.
2020-03-17 19:18:03,759-INFO: Start evaluate...
2020-03-17 19:18:04,487-INFO: Accumulating evaluatation results...
2020-03-17 19:18:04,535-INFO: mAP(0.50, 11point) = 24.81

从这里的log来看，已经正确地执行了eval任务，请确认一下eval任务是否已经执行完了？

terminate called without an active exception
W0317 19:18:04.659145   628 init.cc:209] Warning: PaddlePaddle catches a failure signal, it may not work properly
W0317 19:18:04.659191   628 init.cc:211] You could check whether you killed PaddlePaddle thread/process accidentally or report the case to PaddlePaddle

这条提示来看，像是退出异常。请问这个错误能稳定复现吗？

Xreki avatar Mar 18 '20 03:03 Xreki

重复了几次，每次都报同样的错误

在 2020-03-18 11:30:07，"Yiqun Liu" [email protected] 写道：

2020-03-17 19:17:56,902-INFO: Test iter 0

2020-03-17 19:18:03,758-INFO: Test finish iter 69

2020-03-17 19:18:03,759-INFO: Total number of images: 546, inference time: 73.04237257565066 fps.

2020-03-17 19:18:03,759-INFO: Start evaluate...

2020-03-17 19:18:04,487-INFO: Accumulating evaluatation results...

2020-03-17 19:18:04,535-INFO: mAP(0.50, 11point) = 24.81

从这里的log来看，已经正确地执行了eval任务，请确认一下eval任务是否已经执行完了？

terminate called without an active exception

W0317 19:18:04.659145 628 init.cc:209] Warning: PaddlePaddle catches a failure signal, it may not work properly

W0317 19:18:04.659191 628 init.cc:211] You could check whether you killed PaddlePaddle thread/process accidentally or report the case to PaddlePaddle

这条提示来看，像是退出异常。请问这个错误能稳定复现吗？

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

COST-97 avatar Mar 18 '20 07:03 COST-97