DeepFaceLab2.0 ：超详细入门教程!

Yo站长，问个问题啊
train了很多个模型了，但遇到个两难：
一方面每次从头开始train时间太长；另一方面，一个model连续使用很多次了后（我的src一直没变，dst变了大概十来次），特别容易半路崩掉。
所以想问下，能不能直接把model里面的xxx_decoder_src.h5给覆盖到新model里面？试过效果不好，是不是还需要改些什么？

shift+ > 键就可以全部合成了。

2。0版本有很明顯的邊框，閃爍很厲害，顏色怪怪的。調了很多項結果不是很好

站长站长，我训练模型一开始没有问题，第二次训练时候就开始报错
Error: No training data provided.
Traceback (most recent call last):
File “G:\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\mainscripts\Trainer.py”, line 57, in trainerThread
debug=debug,
File “G:\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\models\ModelBase.py”, line 173, in __init__
self.on_initialize()
File “G:\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\models\Model_SAEHD\Model.py”, line 763, in on_initialize
generators_count=dst_generators_count )
File “G:\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\samplelib\SampleGeneratorFace.py”, line 46, in __init__
raise ValueError(‘No training data provided.’)
ValueError: No training data provided.
这种怎么解决

站长你好，我在图片换脸后发现merged文件夹里图片还是原来的，并没有换脸，之前训练模型等步骤都没有问题，这是什么原因呢？希望您看到后能解答下，谢谢~

4) data_src extract full_face S3FD

Traceback (most recent call last):
File “C:\Users\Lenovo\Downloads\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\main.py”, line 349, in
arguments.func(arguments)
File “C:\Users\Lenovo\Downloads\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\main.py”, line 41, in process_extract
force_gpu_idxs = [ int(x) for x in arguments.force_gpu_idxs.split(‘,’) ] if arguments.force_gpu_idxs is not None else None,
File “C:\Users\Lenovo\Downloads\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\mainscripts\Extractor.py”, line 703, in main
if not cpu_only else nn.DeviceConfig.CPU()
File “C:\Users\Lenovo\Downloads\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\core\leras\nn.py”, line 258, in ask_choose_device_idxs
devices = Devices.getDevices()
File “C:\Users\Lenovo\Downloads\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\core\leras\device.py”, line 144, in getDevices
raise Exception(“nn devices are not initialized. Run initialize_main_env() in main process.”)
Exception: nn devices are not initialized. Run initialize_main_env() in main process.
Pressione qualquer tecla para continuar. . .

为什么我双击8)merged to mp4后，输出完的还是原来的视频?

7) merge Quick96.bat
done 之后还能修改merge merge 的图吗
根本假到一个恶心，参照 deepfaker.xyz 不能用个链结或文章标题吗

站长你好！请问一下我用“train Quick96” 跑到“Initializing models: 100%”之后一直不见出现有您在文章中所说“③中的一行一行跳动的数字”的情况，就一直在“Initializing models: 100%”不动，请问这是卡住了还是它一直在训练中？

我想问一下，我训练出的模型，能不能够复用呢，用在其他的替换上？

请问7）第七步没有出现交互式窗口是怎么回事呢？

站长你好，第一次在训练模型时，loss值在0. 15左右就基本不变了，请问是否和原视频的清晰度有关（src那个视频很糊），还是其它什么问题

站长您好。请问在dst视频拍摄前期加上动捕常用的标记点（在脸上画一些黑色小点）。会对我们的训练和最终效果有帮助或者提高吗？

按回车也没反应
Initializing models: 0%| | 0/5 [00:00<?, ?it/s]
Error: Cannot assign a device for operation encoder/down1/downs_0/conv1/weight/Initializer/_cai_: Could not satisfy explicit device specification '' because the node node encoder/down1/downs_0/conv1/weight/Initializer/_cai_ (defined at D:\Deepfacelab\DeepFaceLab_NVIDIA\_internal\DeepFaceLab\core\leras\initializers\__init__.py:13) having device Device assignments active during op 'encoder/down1/downs_0/conv1/weight/Initializer/_cai_' creation:
with tf.device(None):
with tf.device(/GPU:0): was colocated with a group of nodes that required incompatible device ‘/device:GPU:0’
Colocation Debug Info:
Colocation group had the following types and devices:
Assign: CPU
Const: CPU
Fill: CPU
VariableV2: CPU
Identity: CPU

Colocation members and user-requested devices:
encoder/down1/downs_0/conv1/weight/Initializer/_cai_/shape_as_tensor (Const)
encoder/down1/downs_0/conv1/weight/Initializer/_cai_/Const (Const)
encoder/down1/downs_0/conv1/weight/Initializer/_cai_ (Fill)
encoder/down1/downs_0/conv1/weight (VariableV2) /device:GPU:0
encoder/down1/downs_0/conv1/weight/Assign (Assign) /device:GPU:0
encoder/down1/downs_0/conv1/weight/read (Identity) /device:GPU:0
Assign_1 (Assign) /device:GPU:0
Assign_172 (Assign) /device:GPU:0