deepfacelab使用说明（二）

你好，前面都很顺利，我在训练模型时开始不断出现错误，先是几个程序错误，比如0xc000012d或者0xc0000142，其他软件开始不断崩溃，修复重启后好点，又出现了OpenCV Error: Unspecified error (The numpy array of typenum=11, ndims=3 can not be created) in file 。。。。。。cv2.cpp, line 181
不知道是哪里的问题

usage: main.py [-h] [–tf-suppress-std] {extract,sort,train,convert} …
main.py: error: unrecognized arguments: –ask-for-params
请按任意键继续. .

Running trainer.

Loading model…
Error: Sorry, this model works only on 2GB+ GPU
Traceback (most recent call last):
File “E:\1\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Trainer.py”, line 41, in trainerThread
**in_options)
File “E:\1\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\ModelBase.py”, line 108, in __init__
self.onInitialize(**in_options)
File “E:\1\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\Model_H128\Model.py”, line 23, in onInitialize
self.set_vram_batch_requirements( {2.5:2,3:2,4:2,4:4,5:8,6:8,7:16,8:16,9:24,10:24,11:32,12:32,13:48} )
File “E:\1\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\ModelBase.py”, line 323, in set_vram_batch_requirements
raise Exception (‘Sorry, this model works only on %dGB+ GPU’ % ( keys[0] ) )
Exception: Sorry, this model works only on 2GB+ GPU
请按任意键继续. . .
显存不足 2gb 的怎么办啊 deepfacelib

求解啊

有大佬教一下怎么办吗
第7步转换47%了然后突然出现这个
Exception while process data [undefined]: Traceback (most recent call last):
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\utils\SubprocessorBase.py”, line 233, in subprocess
result = self.onClientProcessData (data)
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 170, in onClientProcessData
image = self.converter.convert_face(image, image_landmarks, self.debug)
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\ConverterMasked.py”, line 179, in convert_face
out_img = np.clip( img_bgr*(1-img_face_mask_aaa) + (out_img*img_face_mask_aaa) , 0, 1.0 )
MemoryError
Exception while process data [undefined]: Traceback (most recent call last):
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\utils\SubprocessorBase.py”, line 233, in subprocess
result = self.onClientProcessData (data)
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 170, in onClientProcessData
image = self.converter.convert_face(image, image_landmarks, self.debug)
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\ConverterMasked.py”, line 179, in convert_face
out_img = np.clip( img_bgr*(1-img_face_mask_aaa) + (out_img*img_face_mask_aaa) , 0, 1.0 )
MemoryError

Exception while process data [undefined]: Traceback (most recent call last):
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\utils\SubprocessorBase.py”, line 233, in subprocess
result = self.onClientProcessData (data)
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 170, in onClientProcessData
image = self.converter.convert_face(image, image_landmarks, self.debug)
File “D:\DeepFaceLab_build_02_12_2018\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\ConverterMasked.py”, line 179, in convert_face
out_img = np.clip( img_bgr*(1-img_face_mask_aaa) + (out_img*img_face_mask_aaa) , 0, 1.0 )
MemoryError

训练的时候，为什么显卡都没有负载，主要是cpu在工作啊？我2g的显卡，预览才显示2行

全程研究好了自己试了一次1分钟的视频转换，把目标视频截取脸后的相似的都删除以后，第七步转化时候有很多现实no face found，找不到脸（全帧截图未删除），第八步整合视频后发现脸99%没有换，只是脸部会闪。不知道何原因？2G显卡用的H64，真心请版主指点，感谢为盼！

大神求助啊，h128模型换脸出下面的错为什么啊
Traceback (most recent call last):
File “”, line 1, in
File “multiprocessing\spawn.py”, line 105, in spawn_main
Error: [Errno 32] Broken pipe
File “multiprocessing\spawn.py”, line 115, in _main
File “F:\ruanjian\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 7, in
from utils import image_utils
File “F:\ruanjian\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\utils\image_utils.py”, line 6, in
Traceback (most recent call last):
File “F:\ruanjian\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 261, in main
**in_options ).process()
File “F:\ruanjian\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\utils\SubprocessorBase.py”, line 100, in process
p.start()
File “multiprocessing\process.py”, line 105, in start
File “multiprocessing\context.py”, line 223, in _Popen
File “multiprocessing\context.py”, line 322, in _Popen
File “multiprocessing\popen_spawn_win32.py”, line 65, in __init__
File “multiprocessing\reduction.py”, line 60, in dump
BrokenPipeError: [Errno 32] Broken pipe
from scipy.spatial import Delaunay
File “F:\ruanjian\DeepFaceLabTorrent\_internal\bin\lib\site-packages\scipy\spatial\__init__.py”, line 96, in
from .qhull import *
ImportError: DLL load failed: 页面文件太小，无法完成操作。

想问问如上操作之后merged的脸没有变 。 。。 是因为训练不足吗 还是我可能有什么地方错误操作

这是什么原因啊，用的H128
===== Model summary =====
== Model name: H128
==
== Current epoch: 0
==
== Options:
== |== batch_size : 16
== |== multi_gpu : False
== |== created_vram_gb : 8.0
== Running on:
== |== [0 : GeForce GTX 1070]
=========================
Saving…
请按任意键继续. . .

== |== batch_size : 8
== |== multi_gpu : False
== |== created_vram_gb : 5.0
== Running on:
== |== [0 : GeForce GTX 1060 5GB]
=========================
Saving…
Starting. Press “Enter” to stop training and save model.
请按任意键继续. . .
没弹出窗口是怎么回事呀

在执行7) H64模型换脸操作的时候报错
no faces found for 10476.png, copying without faces
Converting: 100%|███████████████████████████████████████████████████ ████████████| 10477/10477 [00:29<00:00, 355.21it/s]
Exception ignored in: <bound method BaseSession._Callable.__del__ of >
Traceback (most recent call last):
File “G:\DeepFaceLabTorrent\_internal\bin\lib\site-packages\tensorflow\python\client\session.py”, line 1413, in __del__
AttributeError: ‘NoneType’ object has no attribute ‘raise_exception_on_not_ok_status’

Error: [Errno 32] Broken pipe
Traceback (most recent call last):
File “”, line 1, in
File “multiprocessing\spawn.py”, line 105, in spawn_main
File “multiprocessing\spawn.py”, line 115, in _main
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 12, in
from models import ConverterBase
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\__init__.py”, line 4, in
from .ModelBase import ModelBase
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\models\ModelBase.py”, line 12, in
Traceback (most recent call last):
import gpufmkmgr
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 261, in main
**in_options ).process()
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\gpufmkmgr\__init__.py”, line 1, in
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\utils\SubprocessorBase.py”, line 100, in process
p.start()
File “multiprocessing\process.py”, line 105, in start
from .gpufmkmgr import *
File “multiprocessing\context.py”, line 223, in _Popen
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\gpufmkmgr\gpufmkmgr.py”, line 306, in
File “multiprocessing\context.py”, line 322, in _Popen
File “multiprocessing\popen_spawn_win32.py”, line 65, in __init__
File “multiprocessing\reduction.py”, line 60, in dump
prefer_GPUConfig = GPUConfig()
BrokenPipeError: [Errno 32] Broken pipe
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\gpufmkmgr\gpufmkmgr.py”, line 292, in __init__
gpu_idx = force_best_gpu_idx if (force_best_gpu_idx >= 0 and isValidDeviceIdx(force_best_gpu_idx)) else getBestDeviceIdx() if not choose_worst_gpu else getWorstDeviceIdx()
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\gpufmkmgr\gpufmkmgr.py”, line 215, in getBestDeviceIdx
memInfo = nvmlDeviceGetMemoryInfo( handle )
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\gpufmkmgr\pynvml.py”, line 1154, in nvmlDeviceGetMemoryInfo
_nvmlCheckReturn(ret)
File “D:\DeepFaceLabTorrent\_internal\bin\DeepFaceLab\gpufmkmgr\pynvml.py”, line 310, in _nvmlCheckReturn
raise NVMLError(ret)
gpufmkmgr.pynvml.NVMLError_Unknown: Unknown Error
大佬这个怎么解决

rtx2080
单显卡dlib库提取data-src人脸
显示You have no capable GPUs. Try to close program which can consume VRAM,and run again
如何解决呢
谢谢大佬

Choose mode: (1) hist match, (2) hist match bw, (3) seamless (default), (4) seamless hist match : 3
Choose erode mask modifier [-100..100] (default 0) : 0
Choose blur mask modifier [-100..200] (default 0) : 0
Choose output face scale modifier [-50..50] (default 0) : 0
Transfer color from original DST image? [0..1] (default 0) : 0
Degrade color power of final image [0..100] (default 0) : 0
Export png with alpha channel? [0..1] (default 0) : 0
Running converter.

Loading model…
Error: Dimension 1 in both shapes must be equal, but are 256 and 512. Shapes are [65536,256] and [65536,512]. for ‘Assign_8’ (op: ‘Assign’) with input shapes: [65536,256], [65536,512].
Traceback (most recent call last):
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\framework\ops.py”, line 1626, in _create_c_op
c_op = c_api.TF_FinishOperation(op_desc)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimension 1 in both shapes must be equal, but are 256 and 512. Shapes are [65536,256] and [65536,512]. for ‘Assign_8’ (op: ‘Assign’) with input shapes: [65536,256], [65536,512].

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\DeepFaceLab\mainscripts\Converter.py”, line 38, in model_process
model = models.import_model(model_name)(model_path, **in_options)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\DeepFaceLab\models\ModelBase.py”, line 90, in __init__
self.onInitialize(**in_options)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\DeepFaceLab\models\Model_H128\Model.py”, line 27, in onInitialize
self.encoder.load_weights (self.get_strpath_storage_for_file(self.encoderH5))
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\keras\engine\network.py”, line 1166, in load_weights
f, self.layers, reshape=reshape)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\keras\engine\saving.py”, line 1058, in load_weights_from_hdf5_group
K.batch_set_value(weight_value_tuples)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\keras\backend\tensorflow_backend.py”, line 2465, in batch_set_value
assign_op = x.assign(assign_placeholder)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\ops\variables.py”, line 1493, in assign
name=name)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\ops\state_ops.py”, line 221, in assign
validate_shape=validate_shape)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\ops\gen_state_ops.py”, line 65, in assign
use_locking=use_locking, name=name)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\framework\op_def_library.py”, line 787, in _apply_op_helper
op_def=op_def)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\util\deprecation.py”, line 488, in new_func
return func(*args, **kwargs)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\framework\ops.py”, line 3272, in create_op
op_def=op_def)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\framework\ops.py”, line 1790, in __init__
control_input_ops)
File “F:\BaiduYunDownload\aa\deepfackslib\deepfakes_chinese\_internal\bin\lib\site-packages\tensorflow\python\framework\ops.py”, line 1629, in _create_c_op
raise ValueError(str(e))
ValueError: Dimension 1 in both shapes must be equal, but are 256 and 512. Shapes are [65536,256] and [65536,512]. for ‘Assign_8’ (op: ‘Assign’) with input shapes: [65536,256], [65536,512].

合成的时候显示MEMORY ERROR,我寻思我1050TI的4G显存，8G运行内存，不至于吧