drunkar/dockerfiles

Dockerfiles for dockerhub automated build.

dockerfiles's Issues

Check failed: error == cudaSuccess (38 vs. 0) no CUDA-capable device is detected

I have successfully installed this docker image on my computer which has a Titan X GPU. Also I have installed CUDA toolkit.

➜  ~ ls -la /dev | grep nvidia
crw-rw-rw-   1 root root      195,   0  7月 15 22:06 nvidia0
crw-rw-rw-   1 root root      195, 255  7月 15 22:06 nvidiactl
crw-rw-rw-   1 root root      246,   0  7月 20 11:14 nvidia-uvm

I run this image by the command given.

sudo docker run --name caffe -v /media/ksnzh/DATA/caffe-lstm:/workspace -p 8888:8888 -d --device /dev/nvidia0:/dev/nvidia0 --device /dev/nvidiactl:/dev/nvidiactl --device /dev/nvidia-uvm:/dev/nvidia-uvm drunkar/cuda-caffe-lstm-anaconda /bin/bash

sudo docker exec -it caffe bash

But when I try the coco_caption example, it had some errors.

I0720 12:28:20.358991    47 caffe.cpp:113] Use GPU with device ID 0
F0720 12:28:20.365886    47 common.cpp:131] Check failed: error == cudaSuccess (38 vs. 0)  no CUDA-capable device is detected
*** Check failure stack trace: ***
    @     0x7f752f8819fd  google::LogMessage::Fail()
    @     0x7f752f88389d  google::LogMessage::SendToLog()
    @     0x7f752f8815ec  google::LogMessage::Flush()
    @     0x7f752f8841be  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f752fd01596  caffe::Caffe::SetDevice()
    @           0x4066b2  train()
    @           0x404bd1  main
    @     0x7f752efb8ec5  (unknown)
    @           0x40524c  (unknown)