SAEHD训练不了卡住了，求帮助看看是什么原因

371951625

星级打分

• 1
• 2
• 3
• 4
• 5

平均分:NAN  参与人数:0  我的评分:未评

初始化模型...: 100%|#####################################################################| 5/5 [00:02<00:00,  2.24it/s]
从 D:\AIliandan\0906-RTX30\_internal\pretrain_faces 加载了 15843 个打包的面孔
Sort by yaw: 100%|##################################################################| 128/128 [00:00<00:00, 336.43it/s]
Sort by yaw: 100%|##################################################################| 128/128 [00:00<00:00, 265.35it/s]
/!\ Exception occured in cv2_imread : Traceback (most recent call last):
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\core\cv2ex.py", line 14, in cv2_imread
    bytes = bytearray(loader_func(filename))
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\samplelib\Sample.py", line 106, in read_raw_file
    return f.read (size)
MemoryError
电脑配是华硕天选2plus，显卡RTX3060，下的是汉化之猫版本的30系0906，虚拟内存改到30G了，也开了GPU计划，可以运行训练Quick96，但是用不了训练SAEHD，求解

deeperblue

每天看贴学一点

371951625

During handling of the above exception, another exception occurred:

cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\core\src\alloc.cpp:55: error: (-4:Insufficient memory) Failed to allocate 1048576 bytes in function 'cv::OutOfMemoryError'

Traceback (most recent call last):

During handling of the above exception, another exception occurred:

  File "multiprocessing\process.py", line 258, in _bootstrap
Traceback (most recent call last):
  File "multiprocessing\process.py", line 93, in run
  File "multiprocessing\process.py", line 258, in _bootstrap
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\core\joblib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
  File "multiprocessing\process.py", line 93, in run
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\samplelib\SampleGeneratorFace.py", line 136, in batch_func
    raise Exception ("示例 %s 中出现异常。 错误：%s" % (sample.filename, traceback.format_exc() ) )
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\core\joblib\SubprocessGenerator.py", line 54, in process_func
    gen_data = next (self.generator_func)
Exception: 示例 D:\AIliandan\0906-RTX30\workspace\data_src\aligned\00291.jpg 中出现异常。 错误：Traceback (most recent call last):
  File "D:\AIliandan\0906-RTX30\_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:\AIliandan\0906-RTX30\_internal\DeepFaceLab\samplelib\SampleProcessor.py", line 149, in process
    img = get_eyes_mouth_mask()*mask
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\samplelib\SampleProcessor.py", line 77, in get_eyes_mouth_mask
    mouth_mask = LandmarksProcessor.get_image_mouth_mask (sample_bgr.shape, sample_landmarks)
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\facelib\LandmarksProcessor.py", line 449, in get_image_mouth_mask
    hull_mask = cv2.dilate(hull_mask, cv2.getStructuringElement(cv2.MORPH_ELLIPSE,(dilate,dilate)), iterations = 1 )
cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\core\src\alloc.cpp:55: error: (-4:Insufficient memory) Failed to allocate 1048576 bytes in function 'cv::OutOfMemoryError'


  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\samplelib\SampleGeneratorFace.py", line 136, in batch_func
    raise Exception ("示例 %s 中出现异常。 错误：%s" % (sample.filename, traceback.format_exc() ) )
Exception: 示例 D:\AIliandan\0906-RTX30\workspace\data_src\aligned\00521.jpg 中出现异常。 错误：Traceback (most recent call last):
  File "D:\AIliandan\0906-RTX30\_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:\AIliandan\0906-RTX30\_internal\DeepFaceLab\samplelib\SampleProcessor.py", line 149, in process
    img = get_eyes_mouth_mask()*mask
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\samplelib\SampleProcessor.py", line 76, in get_eyes_mouth_mask
    eyes_mask = LandmarksProcessor.get_image_eye_mask (sample_bgr.shape, sample_landmarks)
  File "D:\AIliandan\0906-RTX30\_internal\DeepFaceLab\facelib\LandmarksProcessor.py", line 427, in get_image_eye_mask
    hull_mask = cv2.dilate(hull_mask, cv2.getStructuringElement(cv2.MORPH_ELLIPSE,(dilate,dilate)), iterations = 1 )
cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\core\src\alloc.cpp:55: error: (-4:Insufficient memory) Failed to allocate 1048576 bytes in function 'cv::OutOfMemoryError'
问题多种多样极度影响炼丹

vancheap

虚拟内存设置为自动试试

371951625

vancheap 发表于 2021-12-8 11:43
虚拟内存设置为自动试试

虚拟内存自动是13个G左右还没手动设置的大呢

371951625

371951625 发表于 2021-12-8 11:56
虚拟内存自动是13个G左右还没手动设置的大呢

我又把虚拟内存增加了，没想到居然行了

guoruoxuan

是的 虚拟内存可以给个60g以上

vancheap

371951625 发表于 2021-12-8 11:56
虚拟内存自动是13个G左右还没手动设置的大呢

是和图片的分辨率有关系，设置为自动方便一些

371951625

vancheap 发表于 2021-12-8 12:19
是和图片的分辨率有关系，设置为自动方便一些

都是默认的参数了

lightkingboy

有点复杂，帮顶上来，别沉。多获得经验

Leon933

進來學習學習，或許日後也會遇上這種問題！