LiveTalking
LiveTalking copied to clipboard
使用docker训练ER-NERF报错
作者您好,我在您提供的docker中尝试训练ER-NERF模型,但是在预处理训练视频的face tracking阶段,出现了以下错误:
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
然后我检查了容器中pytorch的版本是1.11,不是原作者指定的1.12,请问这个错误是pytorch版本不正确导致的吗? 而且我在将pytorch版本改成1.12之后,pytorch3d的运行也会报错: ImportError: /root/miniconda3/lib/python3.8/site-packages/pytorch3d/_C.cpython-38-x86_64-linux-gnu.so: undefined symbol: _ZNK3c1010TensorImpl36is_contiguous_nondefault_policy_implENS_12MemoryFormatE
期待您的回复!