conda环境下Torch not compiled with CUDA enabled解决方法

最新推荐文章于 2026-05-12 09:07:00 发布

原创

最新推荐文章于 2026-05-12 09:07:00 发布 · 2.2k 阅读

标签

#深度学习 #pytorch #人工智能

收录于

1 问题描述

在运行wav2lip模型训练时，报如下错误：

Traceback (most recent call last):
  File "D:\ml\Wav2Lip\preprocess.py", line 32, in <module>
    fa = [face_detection.FaceAlignment(face_detection.LandmarksType._2D, flip_input=False, 
  File "D:\ml\Wav2Lip\preprocess.py", line 32, in <listcomp>
    fa = [face_detection.FaceAlignment(face_detection.LandmarksType._2D, flip_input=False, 
  File "D:\ml\Wav2Lip\face_detection\api.py", line 62, in __init__
    self.face_detector = face_detector_module.FaceDetector(device=device, verbose=verbose)
  File "D:\ml\Wav2Lip\face_detection\detection\sfd\sfd_detector.py", line 28, in __init__
    self.face_detector.to(device)
  File "D:\.conda\wav2lip\lib\site-packages\torch\nn\modules\module.py", line 987, in to
    return self._apply(convert)
  File "D:\.conda\wav2lip\lib\site-packages\torch\nn\modules\module.py", line 639, in _apply
    module._apply(fn)
  File "D:\.conda\wav2lip\lib\site-packages\torch\nn\modules\module.py",