Check failed: error == cudaSuccess (38 vs. 0) no CUDA-capable device is detected
felix-liuying opened this issue · 8 comments
Hello, I have compiled caffe-pro successfully.My cuda 9.1 is ok as follows!
felix@felix-go:~/channel-pruning$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2017 NVIDIA Corporation
Built on Fri_Nov__3_21:07:56_CDT_2017
Cuda compilation tools, release 9.1, V9.1.85
But when I run python3 train.py -action c3 -caffe [GPU0] ,the error appear as follows!
Look forward to your reply! Thank you!
What steps reproduce the bug?
felix@felix-go:~/channel-pruning$ python3 train.py -action c3 -caffe [GPU0]
no lighting pack
WARNING: Logging before InitGoogleLogging() is written to STDERR
F0303 09:50:12.085932 3744 common.cpp:152] Check failed: error == cudaSuccess (38 vs. 0) no CUDA-capable device is detected
What hardware and operating system/distribution are you running?
Operating system: Ubuntu 16.04.3 64bit
CUDA version: 9.1
CUDNN version: v7
openCV version:
BLAS:
Python version: 3.5
If the bug is a crash, provide the backtrace.
@young-fire OK . let me add your WeChatID. Thanks
replace [GPU0] by 0
I have the same problem, have you solved it ?
I have the same problem, have you solved it ?
if replace [GPU0] by 0，If you change, this problem will occur: F0822 17:45:49.191539 12949 image_data_layer.cpp:53] Check failed: !lines_.empty() File is empty
*** Check failure stack trace: ***.
Have you encountered this problem? Or do you know why?thank you very much @wangjx123
@sunny5555 The message
F0822 17:45:49.191539 12949 image_data_layer.cpp:53] Check failed: !lines_.empty() File is empty
has nothing to do with your GPU device name.
The repo maintainer addressed this issue here. According to him
This means you didn't set image data path or mean image path.
Check your temp/vgg.prototxt file (line 12 and 36)
-caffe 0