lipku
wav2lip 制作模特为横屏时，出现张量不一致报错。
各位大佬好： 小弟有个问题，我采样了5分钟的数字人视频，用的是横屏，尺寸为1280720，人脸部截取是384384，然后用wav2lip重新建模后，放入系统data文件夹中，启动后，在提出问题后，出现以下错误代码： Exception in thread Thread-4 (inference): Traceback (most recent call last): File "D:\ProgramData\anaconda3\envs\metahuman\lib\threading.py", line 1016, in _bootstrap_inner self.run() File "D:\ProgramData\anaconda3\envs\metahuman\lib\threading.py", line 953, in run self._target(*self._args, **self._kwargs) File "D:\metahumanstream20250105\lipreal.py", line 150, in inference pred = model(mel_batch, img_batch) File "D:\ProgramData\anaconda3\envs\metahuman\lib\site-packages\torch\nn\modules\module.py", line 1736, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "D:\ProgramData\anaconda3\envs\metahuman\lib\site-packages\torch\nn\modules\module.py", line 1747, in _call_impl return forward_call(*args, **kwargs) File "D:\metahumanstream20250105\wav2lip\models\wav2lip_v2.py", line 150, in forward raise e File "D:\metahumanstream20250105\wav2lip\models\wav2lip_v2.py", line 146, in forward x = torch.cat((x, feats[-1]), dim=1) RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 1 but got size 3 for tensor number 1 in the list.
显示张量大小不一致，诚恳请教我改如何修改。非常感谢。
