Skip to content

Avalanche unable to use cuda device. #1530

@jodie-kang

Description

@jodie-kang

🐛 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
image

🦋 Additional context
NVIDIA-SMI 495.46 Driver Version: 495.46 CUDA Version: 11.5
image

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions