用猫之汉化版本，训练的时候出现这个提示，是电脑哪里.....

我想做琉璃

星级打分

• 1
• 2
• 3
• 4
• 5

平均分:NAN  参与人数:0  我的评分:未评

请教大神，万分感谢，帮我解决的，送上100灵石
选择一个或多个 GPU 编号（以逗号分隔）.

[CPU] : CPU
  [0] : NVIDIA GeForce RTX 4090

[0] 选择哪些 GPU 索引? :
0

Initializing models: 100%|###############################################################| 5/5 [00:01<00:00,  3.89it/s]
加载样本中...: 100%|################################################################| 314/314 [00:00<00:00, 466.35it/s]
Process Process-12:
Traceback (most recent call last):
  File "D:\deepface\DFL_maozhihanhua_RTX3000\_internal\DeepFaceLab\samplelib\SampleGeneratorFace.py", line 134, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.output_sample_types, self.debug, ct_sample=ct_sample)
  File "D:\deepface\DFL_maozhihanhua_RTX3000\_internal\DeepFaceLab\samplelib\SampleProcessor.py", line 143, in process
    mask = get_full_face_mask().copy()
Process Process-25:
Process Process-17:
Process Process-15:
Process Process-22:
Process Process-21:
Process Process-26:
  File "D:\deepface\DFL_maozhihanhua_RTX3000\_internal\DeepFaceLab\samplelib\SampleProcessor.py", line 62, in get_full_face_mask
    xseg_mask = sample.get_xseg_mask()
Traceback (most recent call last):
fatal   Process Process-13:
Traceback (most recent call last):
Traceback (most recent call last):
Traceback (most recent call last):
Traceback (most recent call last):
Process Process-19:
Process Process-24:
Process Process-23:
Traceback (most recent call last):
Traceback (most recent call last):
  File "D:\deepface\DFL_maozhihanhua_RTX3000\_internal\DeepFaceLab\samplelib\Sample.py", line 90, in get_xseg_mask
    return xseg_mask.astype(np.float32) / 255.0
Traceback (most recent call last):
  File "D:\deepface\DFL_maozhihanhua_RTX3000\_internal\DeepFaceLab\samplelib\SampleGeneratorFace.py", line 134, in batch_func
    x, = SampleProcessor.process ([sample], self.sample_process_options, self.output_sample_types, self.debug, ct_sample=ct_sample)
:   File "<string>", line 1, in <mo

3039776685

应该是显存不够或者调大虚拟内存

Deannnn

训练？
你试试把遮罩删掉看看能不能跑
line 143, in process
    mask = get_full_face_mask().copy()
line 62, in get_full_face_mask
    xseg_mask = sample.get_xseg_mask()
line 90, in get_xseg_mask
    return xseg_mask.astype(np.float32) / 255.0
这三条报错都是遮罩出问题，反正我是没遇到过

番茄哥

应该是遮罩问题，换套遮罩试试看

lknet

随便找一张图片,提脸1张,放到dst和src脸图目录里,最小参数直接开训,还报错就是软件环境问题,不报错就是素材问题.