|
发表于 2021-7-19 11:33:53
|
显示全部楼层
大佬帮看看,这个一个报错Memory Error,但是我的是8GB显存,运行内存16GB,虚拟内存不知道
已经安装了cuda,还是这个问题
Error while processing data: Traceback (most recent call last):
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\joblib\SubprocessorBase.py", line 71, in _subprocess_run
result = self.process_data (data)
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\leras\initializers\CA.py", line 48, in process_data
weights = CAInitializerSubprocessor.generate (shape, dtype)
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\leras\initializers\CA.py", line 29, in generate
x = x + np.transpose(x, (0,1,3,2) ) * (1-np.eye(basis_size))
MemoryError: Unable to allocate 48.4 MiB for an array with shape (2048, 86, 6, 6) and data type float64
加载模型: 60%|############################################4 | 3/5 [00:06<00:04, 2.13s/it]
Error: list indices must be integers or slices, not tuple
Traceback (most recent call last):
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\mainscripts\Trainer.py", line 62, in trainerThread
debug=debug)
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\models\ModelBase.py", line 191, in __init__
self.on_initialize()
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\models\Model_AMP\Model.py", line 612, in on_initialize
model.init_weights()
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\leras\layers\Saveable.py", line 104, in init_weights
nn.init_weights(self.get_weights())
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\leras\ops\__init__.py", line 51, in init_weights
nn.batch_set_value( [*zip(ca_tuples_w, nn.initializers.ca.generate_batch (ca_tuples))] )
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\leras\initializers\__init__.py", line 18, in generate_batch
return CAInitializerSubprocessor (data_list).run()
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\joblib\SubprocessorBase.py", line 257, in run
data = self.get_data(cli.host_dict)
File "G:\DeepFaceLab_NVIDIA_up_to_RTX2080Ti\_internal\DeepFaceLab\core\leras\initializers\CA.py", line 67, in get_data
shape, dtype = self.data_list[idx]
TypeError: list indices must be integers or slices, not tuple |
|