I want to install SAM 2 on M1 Macbook, but with CUDA error

[sangwf]

Does that mean I can't do it right now?

Thanks.

[SimonZeng7108]

https://github.com/SauravMaheshkar/samv2

[ronghanghu]

Hi @sangwf, we have recently made the CUDA extension step optional (in https://github.com/facebookresearch/segment-anything-2/pull/155) as a workaround to this problem.

You can pull the latest code and reinstall via

# run the line below inside the SAM 2 repo
git pull;
pip uninstall -y SAM-2;
rm -f sam2/*.so;
pip install -e ".[demo]"

which allows using SAM 2 without CUDA extension (the results should stay the same in most cases, see INSTALL.md for details).

[albertjo]

@ronghanghu I tried following the steps specified in INSTALL.md for the latest code.

# skip the SAM 2 CUDA extension
SAM2_BUILD_CUDA=0 pip install -e ".[demo]"

The installation is successful but when I call build_sam2_video_predictor, I'm getting an AssertionError: Torch not compiled with CUDA enabled error.

I'm following the steps in "video predictor example" notebook. What could I be doing wrong?

[ronghanghu]

@albertjo do you have a GPU in your machine and installed a CUDA version of PyTorch?

Currently the SAM 2 repo doesn't support CPU-only or other devices yet (we might provide this support later). At this moment, if you don't have a GPU on your machine, you may use a few community adaptation in https://github.com/facebookresearch/segment-anything-2/issues/71#issuecomment-2261295494 or https://github.com/facebookresearch/segment-anything-2/pull/123

[ronghanghu]

Hi @sangwf @albertjo we have recently updated the notebooks to include M1 Chip and MPS examples in https://github.com/facebookresearch/segment-anything-2/pull/192. Please pull the latest code from this repo and try them out. Thanks!