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

centos7.3 gpu Not supported #2829

Closed
foxgaga opened this issue Jul 12, 2017 · 4 comments
Closed

centos7.3 gpu Not supported #2829

foxgaga opened this issue Jul 12, 2017 · 4 comments
Assignees

Comments

@foxgaga
Copy link

foxgaga commented Jul 12, 2017

CentOS Linux release 7.3.1611 (Core)
cuda-8.0
cudnn v5

https://github.com/PaddlePaddle/book/blob/develop/05.recommender_system/train.py

alter

paddle.init(use_gpu=True, gpu_id=1)

output

I0712 21:03:33.468981 1598729 Util.cpp:166] commandline:  --use_gpu=True --gpu_id=1
I0712 21:03:36.654531 1598729 GradientMachine.cpp:85] Initing parameters..
I0712 21:03:36.688349 1598729 GradientMachine.cpp:92] Init parameters done.
F0712 21:03:36.766964 1598729 Matrix.cpp:645] Not supported
*** Check failure stack trace: ***
    @     0x7efe05c6406d  google::LogMessage::Fail()
    @     0x7efe05c65eef  google::LogMessage::SendToLog()
    @     0x7efe05c63be3  google::LogMessage::Flush()
    @     0x7efe05c6680e  google::LogMessageFatal::~LogMessageFatal()
    @     0x7efe05abf586  paddle::GpuMatrix::mul()
    @     0x7efe059856b5  paddle::FullyConnectedLayer::forward()
    @     0x7efe05a02c20  paddle::NeuralNetwork::forward()
    @     0x7efe059f4eb3  paddle::GradientMachine::forwardBackward()
    @     0x7efe05c3fbcb  GradientMachine::forwardBackward()
    @     0x7efe058b74ce  _wrap_GradientMachine_forwardBackward
    @     0x7efe0f442412  PyEval_EvalFrameEx
    @     0x7efe0f4441ce  PyEval_EvalCodeEx
    @     0x7efe0f4431f6  PyEval_EvalFrameEx
    @     0x7efe0f4441ce  PyEval_EvalCodeEx
    @     0x7efe0f4431f6  PyEval_EvalFrameEx
    @     0x7efe0f4441ce  PyEval_EvalCodeEx
    @     0x7efe0f4431f6  PyEval_EvalFrameEx
    @     0x7efe0f443323  PyEval_EvalFrameEx
    @     0x7efe0f4441ce  PyEval_EvalCodeEx
    @     0x7efe0f4442e2  PyEval_EvalCode
    @     0x7efe0f464960  PyRun_FileExFlags
    @     0x7efe0f464b3f  PyRun_SimpleFileExFlags
    @     0x7efe0f47a484  Py_Main
    @     0x7efe0e681b35  __libc_start_main
    @           0x400649  (unknown)
Aborted
@llxxxll
Copy link
Member

llxxxll commented Jul 12, 2017

Single GPU settings
paddle.init(use_gpu=True, gpu_id=0)

@qingqing01
Copy link
Contributor

qingqing01 commented Jul 13, 2017

For the sparse input data, the data can not be passed from Python to C++ when using one GPU. The code is here: https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/py_paddle/dataprovider_converter.py#L178 . I will try to fix it later. Thanks for your question.

@foxgaga foxgaga closed this as completed Jul 16, 2017
@foxgaga foxgaga reopened this Jul 16, 2017
@mical892524
Copy link

I have the same issue, how to fix this.

@Superjomn
Copy link
Contributor

Too long ago, free to reopen it.

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

5 participants