Hello,
It looks like you are correct, and the version of torch you are running is a bit old.
Can you check you torch version? In the dev branch, we are currently working with `torch= 2.1.2+cu121`, which has support for the sm_90 compute capability.
import torch torch.cuda.get_arch_list()
['sm_50', 'sm_60', 'sm_70', 'sm_75', 'sm_80', 'sm_86', 'sm_90']
You will almost certainly need to update your torch version, however I would strongly advise against updating to a newer point version of torch, as the utilities have not been tested against all the newer versions (i.e. the newest 2.4.0 has not been tested).
Best, Jackson
From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of AmirHussein Abdolalizadeh amirhussein.a@gmail.com Date: Wednesday, September 25, 2024 at 5:47 AM To: Freesurfer support list freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Repost: NextBrain Segmentation PyTorch-GPU Compatibility Error
External Email - Use Caution Hi,
I am reposting an issue with NextBrain that I asked about a few months back: MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg76705.htmlhttps://secure-web.cisco.com/1Ep1FaXAlpDcDJk3cT-fDgniEdtHQ5Fy30s9xstmgNRrAEq0GrppV042Ou16KilUjuDl29iwITbxqZbAeCRRh6N5BobSaU2JuyHDOFS8u0Wdh-8WJH9P1Td8AVOtYEFnZ4cCE1_ZR2pbsw8RWzQEIbClgbOv0Q5UdZIFadwqFLRgEb0n9nKcFpqirAQ5QYytqyMj4JV3ni4MLXoxIxCoAUsbxx1Nwnlgae3LA7EXS8A174NlMbhCk9a7hwVMlYbOXOIhpgu9xu9JpP4FLRfzuILIYhhagWJKTryrJAqHgjBdRQhENMvKJmFMkik1f3zd2RrBoVMe9QooJEl6QVbMjpQ/https%3A%2F%2Fwww.mail-archive.com%2Ffreesurfer%40nmr.mgh.harvard.edu%2Fmsg76705.html
"NVIDIA H100 with CUDA capability sm_90 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70 sm_75 sm_80 sm_86."
There is a PyTorch incompatibility with the newer H100 nVidia GPUs. I had the same issue with FastBrain which was resolved by updating their torch dependencies in their release. See the discussion here: MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://github.com/Deep-MI/FastSurfer/issues/557https://secure-web.cisco.com/1iI7S9ydIJfS44SEVzbqqTAOd5_M_tzx6bBulJhDsE0KOdQw5bbMhiStE1AQWUVDv8vV9ZXh-dkBuv_d3Dx6JdUHJYnoU41lplDPYOBr7uuDZi1GwW5aT-7ci_YVlxBe-u8PARzU0gGyODJ0hxbkMEPoS0Yv9Uw4SjypQ6yQyFI-FmDd9H4MUFVQB7WkdRWlZLAmr5hfSHXhFpat5laJ5rQhstYP0TDji6qts_Sden0b9R42Sc9uY0bQc4QFKRlimCI4106VOLFh7iEw9ZbZNkm9LPw5Z4ikanczO-TqQtaIeQVjFYNrbCu0OYRReg_VpyP0h1vw7_K-BXygI4Jwa2g/https%3A%2F%2Fgithub.com%2FDeep-MI%2FFastSurfer%2Fissues%2F557
I think this can be a solution for this issue as well since many HPCs are moving towards these kinds of GPUs.
Best, Amir