I cant run my code on gpu sucessfully where i run my code on cpu is fine


#1

The problem like this:
mxnet.base.MXNetError: [06:53:39] src/storage/storage.cc:114: Check failed: e == cudaSuccess || e == cudaErrorCudartUnloading CUDA: no CUDA-capable device is detected

my cuda version: 9.0
my mxnet install: cu-90
I try run my code on gpu it got this wrong,i cant deal with it .


#2

Which platform are you in? can you list your GPUs with nvidia-smi tool?


#3

thank you for your help
my platform on linux 16.04
nvidia-smi can run :
Mon Dec 17 21:25:29 2018
±----------------------------------------------------------------------------+
| NVIDIA-SMI 384.130 Driver Version: 384.130 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 TITAN X (Pascal) Off | 00000000:AF:00.0 Off | N/A |
| 23% 42C P0 58W / 250W | 0MiB / 12189MiB | 3% Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| No running processes found |
±----------------------------------------------------------------------------+

yesterday i try this code:
W1 = nd.random_normal(shape=(1,3),scale=0.01,ctx=mx.gpu())
it got:
[[-0.01320455 0.00682322 -0.00985838]]
<NDArray 1x3 @gpu(0)>
it can run on gpu sucessful , but i run my net (ssd) it doesnt work on gpu what’s wrong with my code .