RuntimeError: cuda runtime error (100) . The gpu is enabled but still giving error

Question

I am new to google Colab and pyTorch. I am running a pytorch model but it is giving me the Cuda Runtime Error in google Colab. My gpu is enabled on google colab but it is still giving error, The description of gpu is available in the image below. Can anyone please help me out?

torch GPU

THCudaCheck FAIL file=/pytorch/aten/src/THC/THCGeneral.cpp line=47 error=100 : no CUDA-capable device is detected
Traceback (most recent call last):
  File "run.py", line 338, in <module>
    main()
  File "run.py", line 303, in main
    model = model.cuda()
  File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 458, in cuda
    return self._apply(lambda t: t.cuda(device))
  File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 354, in _apply
    module._apply(fn)
  File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 354, in _apply
    module._apply(fn)
  File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 354, in _apply
    module._apply(fn)
  File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 376, in _apply
    param_applied = fn(param)
  File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 458, in <lambda>
    return self._apply(lambda t: t.cuda(device))
  File "/usr/local/lib/python3.6/dist-packages/torch/cuda/__init__.py", line 190, in _lazy_init
    torch._C._cuda_init()
RuntimeError: cuda runtime error (100) : no CUDA-capable device is detected at /pytorch/aten/src/THC/THCGeneral.cpp:47
Read prediction from logs/logs_sparc_editsql/valid_use_predicted_queries_predictions.json

What GPU is installed in the system? What CUDA version is installed? Are any GPUs in the system ignored due to environment variable CUDA_VISIBLE_DEVICES? — njuffa
– njuffa, Commented Sep 27, 2020 at 2:40
@njuffa I am using google colab i.sstatic.net/QfCnL.png . This is the specs of gpu. Also when I am training other pytorch model it is working on the same GPU. — Siddharth Rawal
– Siddharth Rawal, Commented Sep 27, 2020 at 3:49
@njuffa Version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2019 NVIDIA Corporation Built on Sun_Jul_28_19:07:16_PDT_2019 Cuda compilation tools, release 10.1, V10.1.243 Also how to know if gpu is ignored? — Siddharth Rawal
– Siddharth Rawal, Commented Sep 27, 2020 at 4:01

Siddharth Rawal · Accepted Answer · 2020-09-27 22:20:34Z

0

Nvm, I was putting CUDA_VISIBLE_DEVICES to 5. It should be the number of CUDA devices you have.

answered Sep 27, 2020 at 22:20

Siddharth Rawal

11 silver badge1 bronze badge

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

RuntimeError: cuda runtime error (100) . The gpu is enabled but still giving error

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related