RuntimeError: Error building extension 'emd_ext'

[OswaldoBornemann]

How to solve it locally ?

[mashiroyuki02]

Hi, did you solve it? I encouter the same issure

[Wanghc233]

export TORCH_CUDA_ARCH_LIST="7.5" before training

[Philcalab]

export TORCH_CUDA_ARCH_LIST="7.5" before training
May I ask why the version number is 7.5? What determines this?

[Wanghc233]

if not use, please writer export TORCH_CUDA_ARCH_LIST="7.5" in terminal. The "8.0" usually represents the NVIDIA Ampere architecture, which includes a range of GPU models, such as the NVIDIA A100.You can use different numbers to match the version of your GPU.

[Philcalab]

if not use, please writer export TORCH_CUDA_ARCH_LIST="7.5" in terminal. The "8.0" usually represents the NVIDIA Ampere architecture, which includes a range of GPU models, such as the NVIDIA A100.You can use different numbers to match the version of your GPU.
Thank you very much for your answer, but it still doesn't seem to solve my problem.