Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

GPU使用multiBinaryLabelCrossEntropy问题 #1900

Closed
CrossLee1 opened this issue Apr 26, 2017 · 2 comments
Closed

GPU使用multiBinaryLabelCrossEntropy问题 #1900

CrossLee1 opened this issue Apr 26, 2017 · 2 comments
Assignees

Comments

@CrossLee1
Copy link
Contributor

我在GPU下训练,使用multiBinaryLabelCrossEntropy的cost,但是报下面的错误,请问是什么问题呢?

F0426 10:41:39.498152 20984 hl_cuda_device.cc:566] Check failed: cudaSuccess == cudaStat (0 vs. 77) Cuda Error: an illegal memory access was encountered
*** Check failure stack trace: ***
@ 0xff34a0 google::LogMessage::Fail()
@ 0xff33fc google::LogMessage::SendToLog()
@ 0xff2d80 google::LogMessage::Flush()
@ 0xff5e27 google::LogMessageFatal::~LogMessageFatal()
@ 0xfb5375 hl_stream_synchronize()
@ 0xfb9f6f hl_matrix_multi_binary_cross_entropy()
@ 0xd25592 paddle::GpuMatrix::multiBinaryLabelCrossEntropy()
@ 0xae957b paddle::MultiBinaryLabelCrossEntropy::forwardImp()
@ 0xae5d24 paddle::CostLayer::forward()
@ 0xbc37a1 paddle::NeuralNetwork::forward()
@ 0xbad6f8 paddle::TrainerThread::forward()
@ 0xbad3ae paddle::TrainerThread::computeThread()
@ 0xbacee1 _ZZN6paddle13TrainerThread5startEvENKUlvE_clEv
@ 0xbb30f0 _ZNSt12_Bind_simpleIFZN6paddle13TrainerThread5startEvEUlvE_vEE9_M_invokeIIEEEvSt12_Index_tupleIIXspT_EEE
@ 0xbb2e4f _ZNSt12_Bind_simpleIFZN6paddle13TrainerThread5startEvEUlvE_vEEclEv
@ 0xbb2cb0 _ZNSt6thread5_ImplISt12_Bind_simpleIFZN6paddle13TrainerThread5startEvEUlvE_vEEE6_M_runEv
@ 0x7fe6666b52d0 execute_native_thread_routine
@ 0x318b207851 (unknown)
@ 0x318aee767d (unknown)
@ (nil) (unknown)

@hedaoyuan
Copy link
Contributor

I will check this function. And is there any way to reproduce the problem.

@hedaoyuan hedaoyuan mentioned this issue Apr 26, 2017
@hedaoyuan
Copy link
Contributor

This illegal memory access problem may be due to incorrect input. @CrossLee1 #1907 add the inspection of input data, and if (row_col[i] >= dimN) will print like

Error: col_num=1001, i=1, row_col[i]=1072468318
F0426 21:08:35.725378 12051 hl_cuda_device.cc:566] Check failed: cudaSuccess == cudaStat (0 vs. 77) Cuda Error: an illegal memory access was encountered
``

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants