You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Running FastSurfer on our HPC cluster with Nvidia H100 fails with an error shown below. Any comments are appreciated.
Screenshots / Log files
Illegal option --
/venv/lib/python3.10/site-packages/torch/cuda/__init__.py:173: UserWarning:
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.
If you want to use the NVIDIA H100 GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/
warnings.warn(incompatible_device_warn.format(device_name, capability, " ".join(arch_list), device_name))
Traceback (most recent call last):
File "/fastsurfer/FastSurferCNN/run_prediction.py", line 630, in <module>
raise e
File "/fastsurfer/FastSurferCNN/run_prediction.py", line 566, in <module>
pred_data = eval.get_prediction(
File "/fastsurfer/FastSurferCNN/run_prediction.py", line 377, in get_prediction
pred_prob = torch.zeros(shape, **kwargs)
RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.
yes, actually we are finalising a new release, which we are currently delaying a little to add this and also make sure that native install works on the Mac. See our open PRs, specifically #560
Question/Support Request
Running FastSurfer on our HPC cluster with Nvidia H100 fails with an error shown below. Any comments are appreciated.
Screenshots / Log files
Environment
The text was updated successfully, but these errors were encountered: