search

CVHub520 / X-AnyLabeling

Effortless data labeling with AI support from Segment Anything and other awesome models.
GNU General Public License v3.0
4.27k stars 484 forks source link

Error in loading model: segment_anything_2_video with error: Torch not compiled with CUDAA enabled #647

Closed Sade686 closed 2 months ago

Sade686 commented 2 months ago

Search before asking

Question

ERROR | model_manager:_load_model:1046 - ❌ Error in loading model: segment_anything_2_video with error: Torch not compiled with CUDAA enabled

Additional

No response

Sade686 commented 2 months ago

cuda 11.8

Sade686 commented 2 months ago

torch 2.4.1

CVHub520 commented 2 months ago

Hey there! @Sade686,

Thank you for bringing this issue to our attention. The error message you provided indicates that the issue is not related to the model itself, but rather a problem with the configuration of your local PyTorch and CUDA environment.

I would suggest trying the following steps to resolve the problem:

  1. Check the compatibility between the CUDA version installed on your system and the version required by the PyTorch library you are using. You can find the compatibility information online.

  2. Try reinstalling or updating PyTorch and CUDA to ensure they are compatible and properly configured. You can refer to the PyTorch official documentation for installation guidelines.

  3. If the above steps do not work, you can also try running the code in a fresh environment, such as a Docker container or a virtual environment.

Please let me know if you encounter any issues while trying these steps. I'll be happy to provide further assistance.

Best regards, X-AnyLabeling Maintainer