Avalanche unable to use cuda device. #1582
Unanswered
jodie-kang
asked this question in
Q&A
Replies: 3 comments
-
Beta Was this translation helpful? Give feedback.
0 replies
-
Beta Was this translation helpful? Give feedback.
0 replies
-
looks like an environment issue due to installing pytorch with conda and avalanche with pip.
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
🐛 Describe the bug
UserWarning: CUDA initialization: The NVIDIA driver on your system is too old (found version 11050). Please update your GPU driver by downloading and installing a new version from the URL: http://www.nvidia.com/Download/index.aspx Alternatively, go to: https://pytorch.org to install a PyTorch version that has been compiled with your version of the CUDA driver. (Triggered internally at ../c10/cuda/CUDAFunctions.cpp:108.)
🐜 To Reproduce
Installation Process:
Observations: Automatic installation of torch-2.1.0
Attempt 1: Manual installation of the GPU version of PyTorch:
conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.3 -c pytorch
Encountered an error: TypeError: BaseSGDTemplate.init() missing 2 required positional arguments: 'model' and 'optimizer'
Observation: The above command led to automatic uninstallation of torch-2.1.0
Attempt 2: pip install typing-extensions==4.4.0, but the problem persists
Attempt 3: Created a conda environment with Python=3.10, but the issue remains unresolved.
🐝 Expected behavior
Due to certain reasons, I lack permission to change the CUDA version on the server. Are there any alternative methods to address the above issue?
🐞 Screenshots

🦋 Additional context

NVIDIA-SMI 495.46 Driver Version: 495.46 CUDA Version: 11.5
Beta Was this translation helpful? Give feedback.
All reactions