Beeno5900
Reputation: 31
Pytorch cuda is unavailable even installed CUDA and pytorch with cuda. How to fix?
I'm trying to use pytorch with my GPU (RTX 3070) on my Windows machine using WSL2, but I couldn't get it work even though I followed the Nvidia guide (https://docs.nvidia.com/cuda/wsl-user-guide/index.html#abstract).
nvidia-smi.exe output:
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 471.41 Driver Version: 471.41 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+
| GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce ... WDDM | 00000000:0A:00.0 On | N/A |
| 0% 40C P5 12W / 220W | 1815MiB / 8192MiB | 0% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| 0 N/A N/A 1384 C+G Insufficient Permissions N/A |
| 0 N/A N/A 1628 C+G ...dows\System32\WWAHost.exe N/A |
| 0 N/A N/A 3172 C+G ...y\ShellExperienceHost.exe N/A |
| 0 N/A N/A 5940 C+G ...lPanel\SystemSettings.exe N/A |
| 0 N/A N/A 6360 C+G C:\Windows\explorer.exe N/A |
| 0 N/A N/A 7280 C+G ...artMenuExperienceHost.exe N/A |
| 0 N/A N/A 7732 C+G ...5n1h2txyewy\SearchApp.exe N/A |
| 0 N/A N/A 8256 C+G ...ekyb3d8bbwe\YourPhone.exe N/A |
| 0 N/A N/A 8780 C+G ...nputApp\TextInputHost.exe N/A |
| 0 N/A N/A 10032 C+G ...perience\NVIDIA Share.exe N/A |
| 0 N/A N/A 10732 C+G ...hyper\app-3.0.2\Hyper.exe N/A |
| 0 N/A N/A 10804 C+G ...kyb3d8bbwe\Calculator.exe N/A |
| 0 N/A N/A 10852 C+G ...in7x64\steamwebhelper.exe N/A |
| 0 N/A N/A 12180 C+G ...ge\Application\msedge.exe N/A |
| 0 N/A N/A 13880 C+G ...icrosoft VS Code\Code.exe N/A |
| 0 N/A N/A 14724 C+G ...b3d8bbwe\WinStore.App.exe N/A |
+-----------------------------------------------------------------------------+
nvcc -v output:
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Wed_Jun__2_19:15:15_PDT_2021
Cuda compilation tools, release 11.4, V11.4.48
Build cuda_11.4.r11.4/compiler.30033411_0
dpkg -l | grep nvidia output:
ii libnvidia-cfg1-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA binary OpenGL/GLX configuration library
ii libnvidia-common-470 470.57.02-0ubuntu0.20.04.1 all Shared files used by the NVIDIA libraries
rc libnvidia-compute-450:amd64 450.51.05-0ubuntu1 amd64 NVIDIA libcompute package
rc libnvidia-compute-465:amd64 465.27-0ubuntu0.20.04.2 amd64 NVIDIA libcompute package
ii libnvidia-compute-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA libcompute package
ii libnvidia-decode-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA Video Decoding runtime libraries
ii libnvidia-encode-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 NVENC Video Encoding runtime library
ii libnvidia-extra-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 Extra libraries for the NVIDIA driver
ii libnvidia-fbc1-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA OpenGL-based Framebuffer Capture runtime library
ii libnvidia-gl-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA OpenGL/GLX/EGL/GLES GLVND libraries
and Vulkan ICD
ii libnvidia-ifr1-470:amd64 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA OpenGL-based Inband Frame Readback runtime library
ii libnvidia-ml-dev 10.1.243-3 amd64 NVIDIA Management Library (NVML) development files
rc nvidia-compute-utils-450 450.51.05-0ubuntu1 amd64 NVIDIA compute utilities
ii nvidia-compute-utils-470 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA compute utilities
ii nvidia-cuda-dev 10.1.243-3 amd64 NVIDIA CUDA development files
ii nvidia-cuda-doc 10.1.243-3 all NVIDIA CUDA and OpenCL documentation
ii nvidia-cuda-gdb 10.1.243-3 amd64 NVIDIA CUDA Debugger (GDB)
ii nvidia-cuda-toolkit 10.1.243-3 amd64 NVIDIA CUDA development toolkit
rc nvidia-dkms-450 450.51.05-0ubuntu1 amd64 NVIDIA DKMS package
ii nvidia-dkms-470 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA DKMS package
ii nvidia-driver-470 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA driver metapackage
rc nvidia-kernel-common-450 450.51.05-0ubuntu1 amd64 Shared files used with the kernel module
ii nvidia-kernel-common-470 470.57.02-0ubuntu0.20.04.1 amd64 Shared files used with the kernel module
ii nvidia-kernel-source-470 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA kernel source package
ii nvidia-opencl-dev:amd64 10.1.243-3 amd64 NVIDIA OpenCL development files
ii nvidia-prime 0.8.16~0.20.04.1 all Tools to enable NVIDIA's Prime
ii nvidia-profiler 10.1.243-3 amd64 NVIDIA Profiler for CUDA and OpenCL
ii nvidia-settings 450.51.05-0ubuntu1 amd64 Tool for configuring the NVIDIA graphics driver
ii nvidia-utils-470 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA driver support binaries
ii nvidia-visual-profiler 10.1.243-3 amd64 NVIDIA Visual Profiler for CUDA and OpenCL
ii screen-resolution-extra 0.18build1 all Extension for the nvidia-settings control panel
ii xserver-xorg-video-nvidia-470 470.57.02-0ubuntu0.20.04.1 amd64 NVIDIA binary Xorg driver
dpkg -l | grep cuda output:
ii cuda 11.4.0-1 amd64 CUDA meta-package
ii cuda-11-4 11.4.0-1 amd64 CUDA 11.4 meta-package
ii cuda-cccl-11-4 11.4.43-1 amd64 CUDA CCCL
ii cuda-command-line-tools-11-4 11.4.0-1 amd64 CUDA command-line tools
ii cuda-compiler-11-4 11.4.0-1 amd64 CUDA compiler
rc cuda-cudart-11-0 11.0.194-1 amd64 CUDA Runtime native Libraries
ii cuda-cudart-11-4 11.4.43-1 amd64 CUDA Runtime native Libraries
ii cuda-cudart-dev-11-4 11.4.43-1 amd64 CUDA Runtime native dev links, headers
ii cuda-cuobjdump-11-4 11.4.43-1 amd64 CUDA cuobjdump
ii cuda-cupti-11-4 11.4.65-1 amd64 CUDA profiling tools runtime libs.
ii cuda-cupti-dev-11-4 11.4.65-1 amd64 CUDA profiling tools interface.
ii cuda-cuxxfilt-11-4 11.4.43-1 amd64 CUDA cuxxfilt
ii cuda-demo-suite-11-4 11.4.43-1 amd64 Demo suite for CUDA
ii cuda-documentation-11-4 11.4.43-1 amd64 CUDA documentation
ii cuda-driver-dev-11-4 11.4.43-1 amd64 CUDA Driver native dev stub library
ii cuda-gdb-11-4 11.4.55-1 amd64 CUDA-GDB
ii cuda-libraries-11-4 11.4.0-1 amd64 CUDA Libraries 11.4 meta-package
ii cuda-libraries-dev-11-4 11.4.0-1 amd64 CUDA Libraries 11.4 development meta-package
ii cuda-memcheck-11-4 11.4.43-1 amd64 CUDA-MEMCHECK
ii cuda-nsight-11-4 11.4.43-1 amd64 CUDA nsight
ii cuda-nsight-compute-11-4 11.4.0-1 amd64 NVIDIA Nsight Compute
ii cuda-nsight-systems-11-4 11.4.0-1 amd64 NVIDIA Nsight Systems
ii cuda-nvcc-11-4 11.4.48-1 amd64 CUDA nvcc
ii cuda-nvdisasm-11-4 11.4.43-1 amd64 CUDA disassembler
ii cuda-nvml-dev-11-4 11.4.43-1 amd64 NVML native dev links, headers
ii cuda-nvprof-11-4 11.4.43-1 amd64 CUDA Profiler tools
ii cuda-nvprune-11-4 11.4.43-1 amd64 CUDA nvprune
ii cuda-nvrtc-11-4 11.4.50-1 amd64 NVRTC native runtime libraries
ii cuda-nvrtc-dev-11-4 11.4.50-1 amd64 NVRTC native dev links, headers
ii cuda-nvtx-11-4 11.4.43-1 amd64 NVIDIA Tools Extension
ii cuda-nvvp-11-4 11.4.43-1 amd64 CUDA Profiler tools
ii cuda-repo-ubuntu2004-11-0-local 11.0.2-450.51.05-1 amd64 cuda repository configuration files
ii cuda-repo-wsl-ubuntu-11-4-local 11.4.0-1 amd64 cuda repository configuration files
ii cuda-runtime-11-4 11.4.0-1 amd64 CUDA Runtime 11.4 meta-package
ii cuda-samples-11-4 11.4.43-1 amd64 CUDA example applications
ii cuda-sanitizer-11-4 11.4.54-1 amd64 CUDA Sanitizer
rc cuda-toolkit-11-0 11.0.2-1 amd64 CUDA Toolkit 11.0 meta-package
ii cuda-toolkit-11-4 11.4.0-1 amd64 CUDA Toolkit 11.4 meta-package
ii cuda-toolkit-11-4-config-common 11.4.43-1 all Common config package for CUDA Toolkit 11.4.
ii cuda-toolkit-11-config-common 11.4.43-1 all Common config package for CUDA Toolkit 11.
ii cuda-toolkit-config-common 11.4.43-1 all Common config package for CUDA Toolkit.
ii cuda-tools-11-4 11.4.0-1 amd64 CUDA Tools meta-package
rc cuda-visual-tools-11-0 11.0.2-1 amd64 CUDA visual tools
ii cuda-visual-tools-11-4 11.4.0-1 amd64 CUDA visual tools
ii libcudart10.1:amd64 10.1.243-3 amd64 NVIDIA CUDA Runtime Library
ii nvidia-cuda-dev 10.1.243-3 amd64 NVIDIA CUDA development files
ii nvidia-cuda-doc 10.1.243-3 all NVIDIA CUDA and OpenCL documentation
ii nvidia-cuda-gdb 10.1.243-3 amd64 NVIDIA CUDA Debugger (GDB)
ii nvidia-cuda-toolkit 10.1.243-3 amd64 NVIDIA CUDA development toolkit
The pytorch version I installed:
torch 1.9.0+cu111
torchaudio 0.9.0
torchvision 0.10.0+cu111
But pytorch tells me that cuda is not available:
Python 3.8.10 (default, Jun 2 2021, 10:49:15)
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> torch.cuda.is_available()
False
Does anyone know if I installed all the packages I need? I tried many solutions posted on the internet but still doesn't work.
Upvotes: 3
Views: 13065
Answers (3)
Jérémy Talbot-Pâquet
Reputation: 371
For me it worked when I did python -m pip install torch==1.13.1+cu117 -f https://download.pytorch.org/whl/torch_stable.html otherwise, only torchaudio had +cu117
Upvotes: 0
Larry Li
Reputation: 31
My environment is (Ubuntu 20.04 with NVIDIA GTX 1080Ti):
$ nvidia-smi | grep CUDA
| NVIDIA-SMI 470.74 Driver Version: 470.74 CUDA Version: 11.4 |
$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Sun_Aug_15_21:14:11_PDT_2021
Cuda compilation tools, release 11.4, V11.4.120
Build cuda_11.4.r11.4/compiler.30300941_0
After the installation of CUDA and cuDNN, I run this followwing command getting from https://pytorch.org/get-started/locally/ with options (PyTorch Stable 1.9.1, Linux, Pip, Python, CUDA 11.1).
pip3 install torch==1.9.1+cu111 torchvision==0.10.1+cu111 torchaudio==0.9.1 -f https://download.pytorch.org/whl/torch_stable.html
Find packages:
(test-py38) ➜ ~ conda list | grep torch
_pytorch_select 0.1 cpu_0 defaults
pytorch 1.8.1 cpu_py38h60491be_0 defaults
torch 1.9.1+cu111 <pip>
torchaudio 0.9.1 <pip>
torchvision 0.10.1+cu111 <pip>
Then:
$ python
Python 3.8.11 (default, Aug 3 2021, 15:09:35)
[GCC 7.5.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> torch.cuda.is_available()
True
Upvotes: 3
Jon Cory
Reputation: 1
It looks like cudatoolkit 11.4 is installed as currently installed as cudatoolkit-dev:
https://anaconda.org/conda-forge/cudatoolkit-dev
conda install -c conda-forge cudatoolkit-dev
Upvotes: -1
Related Questions
- CUDA initialization: CUDA unknown error - this may be due to an incorrectly set up environment
- Why `torch.cuda.is_available()` returns False even after installing pytorch with cuda?
- Pytorch says that CUDA is not available (on Ubuntu)
- PyTorch Cuda with anaconda not available
- PyTorch having trouble detecting CUDA
- After installing Pytorch cuda , torch.cuda.is_available() show false. What to do?
- cuda is not available on my pytorch, but I can't find anything wrong with the version
- Pytorch doesn't find a CUDA device
- pytorch - net.cuda() seems don't work
- Pytorch not using cuda device