No CUDA capable device is detected

[VibhuSrivastava]

I am running this on a google cloud linux instance (Ubuntu 14.04 trusty LTS)

When I start training of CPU only, it is running fine. I have installed CUDA (7.5) and cuDNN(v5.1 RC) and have followed the given instructions:

• Install Torch.
• luarocks install nn
• luarocks install rnn
• luarocks install penlight
• luarocks install cutorch
• luarocks install cunn

However, when I start training with the help of GPU support, I am receiving the following error:

user1@instance-1:~/neuralconvo$ th train.lua --cuda -- Loading dataset Loading vocabulary from data/vocab.t7 ... Dataset stats: Vocabulary size: 35147 Examples: 221282 THCudaCheck FAIL file=/tmp/luarocks_cutorch-scm-1-515/cutorch/lib/THC/THCGeneral.c line=20 error=38 : no CUDA- capable device is detected /home/user1/torch/install/bin/luajit: ...user1/torch/install/share/lua/5.1/trepl/init.lua:384: cuda runtime error (38) : no CUDA-capable device is detected at /tmp/luarocks_cutorch-scm-1-515/cutorch/lib/T HC/THCGeneral.c:20 stack traceback: [C]: in function 'error' ...user1/torch/install/share/lua/5.1/trepl/init.lua:384: in function 'require' train.lua:55: in main chunk [C]: in function 'dofile' ...dita/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:145: in main chunk [C]: at 0x00406670

What is going wrong and how can I fix this?

[vikram-gupta]

Hey @VibhuSrivastava ,

Were you able to build and run the CUDA samples? What is the output of nvidia-smi? Is the machine having an NVIDIA GPU?