请问：训练模型修改参数时没有发现 true_face_power，为何？

pheobus

星级打分

• 1
• 2
• 3
• 4
• 5

平均分:NAN  参与人数:0  我的评分:未评

请问：前段时间购买了一个预训练模型，继续训练时，想更改一些参数，交互提示时没有发现 提示true_face_power的信息，这是为什么呢？
D:\workroom\0906-RTX30>call _internal\setenv.bat

D:\workroom\0906-RTX30>rem ========== BASE ENV ==========

D:\workroom\0906-RTX30>SET INTERNAL=D:\workroom\0906-RTX30\_internal\

D:\workroom\0906-RTX30>SET INTERNAL=D:\workroom\0906-RTX30\_internal

D:\workroom\0906-RTX30>rem overriding windows user/local environment

D:\workroom\0906-RTX30>SET LOCALENV_DIR=D:\workroom\0906-RTX30\_internal\_e

D:\workroom\0906-RTX30>SET TMP=D:\workroom\0906-RTX30\_internal\_e\t

D:\workroom\0906-RTX30>SET TEMP=D:\workroom\0906-RTX30\_internal\_e\t

D:\workroom\0906-RTX30>SET HOME=D:\workroom\0906-RTX30\_internal\_e\u

D:\workroom\0906-RTX30>SET HOMEPATH=D:\workroom\0906-RTX30\_internal\_e\u

D:\workroom\0906-RTX30>SET USERPROFILE=D:\workroom\0906-RTX30\_internal\_e\u

D:\workroom\0906-RTX30>SET LOCALAPPDATA=D:\workroom\0906-RTX30\_internal\_e\u\AppData\Local

D:\workroom\0906-RTX30>SET APPDATA=D:\workroom\0906-RTX30\_internal\_e\u\AppData\Roaming

D:\workroom\0906-RTX30>rem ========== PYTHON ENV ==========

D:\workroom\0906-RTX30>SET PYTHON_PATH=D:\workroom\0906-RTX30\_internal\python-3.7

D:\workroom\0906-RTX30>rem overriding default python env vars in order not to interfere with any system python installation

D:\workroom\0906-RTX30>SET PYTHONHOME=

D:\workroom\0906-RTX30>SET PYTHONPATH=

D:\workroom\0906-RTX30>SET PYTHONEXECUTABLE=D:\workroom\0906-RTX30\_internal\python-3.7\python.exe

D:\workroom\0906-RTX30>SET PYTHONWEXECUTABLE=D:\workroom\0906-RTX30\_internal\python-3.7\pythonw.exe

D:\workroom\0906-RTX30>SET PYTHON_EXECUTABLE=D:\workroom\0906-RTX30\_internal\python-3.7\python.exe

D:\workroom\0906-RTX30>SET PYTHONW_EXECUTABLE=D:\workroom\0906-RTX30\_internal\python-3.7\pythonw.exe

D:\workroom\0906-RTX30>SET PYTHON_BIN_PATH=D:\workroom\0906-RTX30\_internal\python-3.7\python.exe

D:\workroom\0906-RTX30>SET PYTHON_LIB_PATH=D:\workroom\0906-RTX30\_internal\python-3.7\Lib\site-packages

D:\workroom\0906-RTX30>SET QT_QPA_PLATFORM_PLUGIN_PATH=D:\workroom\0906-RTX30\_internal\python-3.7\Lib\site-packages\PyQt5\Qt\plugins

D:\workroom\0906-RTX30>SET PATH=D:\workroom\0906-RTX30\_internal\python-3.7;D:\workroom\0906-RTX30\_internal\python-3.7\Scripts;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\libnvvp;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Windows\System32\OpenSSH\;F:\Users\Win-10\anaconda3;F:\Program Files\Git\cmd;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\NVIDIA Corporation\NVIDIA NvDLISR;C:\Program Files\NVIDIA Corporation\Nsight Compute 2021.3.1\;C:\Users\Win-10\AppData\Local\Microsoft\WindowsApps;C:\Program Files\Multipass\bin

D:\workroom\0906-RTX30>rem ========== CUDA ENV ==========

D:\workroom\0906-RTX30>SET PATH=D:\workroom\0906-RTX30\_internal\CUDA;D:\workroom\0906-RTX30\_internal\CUDNN;D:\workroom\0906-RTX30\_internal\python-3.7;D:\workroom\0906-RTX30\_internal\python-3.7\Scripts;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\libnvvp;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Windows\System32\OpenSSH\;F:\Users\Win-10\anaconda3;F:\Program Files\Git\cmd;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\NVIDIA Corporation\NVIDIA NvDLISR;C:\Program Files\NVIDIA Corporation\Nsight Compute 2021.3.1\;C:\Users\Win-10\AppData\Local\Microsoft\WindowsApps;C:\Program Files\Multipass\bin

D:\workroom\0906-RTX30>for /F "tokens=4-7 delims=[.] " %i in ('ver') do (if %i == Version (set v=%j.%k )  else (set v=%i.%j ) )

D:\workroom\0906-RTX30>(if 10 == Version (set v=0.19044 )  else (set v=10.0 ) )

D:\workroom\0906-RTX30>if "10.0" == "10.0" (SET "PATH=D:\workroom\0906-RTX30\_internal\CUDNN\Win10.0;D:\workroom\0906-RTX30\_internal\CUDA;D:\workroom\0906-RTX30\_internal\CUDNN;D:\workroom\0906-RTX30\_internal\python-3.7;D:\workroom\0906-RTX30\_internal\python-3.7\Scripts;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\libnvvp;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Windows\System32\OpenSSH\;F:\Users\Win-10\anaconda3;F:\Program Files\Git\cmd;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\NVIDIA Corporation\NVIDIA NvDLISR;C:\Program Files\NVIDIA Corporation\Nsight Compute 2021.3.1\;C:\Users\Win-10\AppData\Local\Microsoft\WindowsApps;C:\Program Files\Multipass\bin" )  else (SET "PATH=D:\workroom\0906-RTX30\_internal\CUDNN\Win6.x;D:\workroom\0906-RTX30\_internal\CUDA;D:\workroom\0906-RTX30\_internal\CUDNN;D:\workroom\0906-RTX30\_internal\python-3.7;D:\workroom\0906-RTX30\_internal\python-3.7\Scripts;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\libnvvp;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Windows\System32\OpenSSH\;F:\Users\Win-10\anaconda3;F:\Program Files\Git\cmd;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\NVIDIA Corporation\NVIDIA NvDLISR;C:\Program Files\NVIDIA Corporation\Nsight Compute 2021.3.1\;C:\Users\Win-10\AppData\Local\Microsoft\WindowsApps;C:\Program Files\Multipass\bin" )

D:\workroom\0906-RTX30>rem ========== ADDITIONAL ENV ==========

D:\workroom\0906-RTX30>SET XNVIEWMP_PATH=D:\workroom\0906-RTX30\_internal\XnViewMP

D:\workroom\0906-RTX30>SET FFMPEG_PATH=D:\workroom\0906-RTX30\_internal\ffmpeg

D:\workroom\0906-RTX30>SET PATH=D:\workroom\0906-RTX30\_internal\XnViewMP;D:\workroom\0906-RTX30\_internal\ffmpeg;D:\workroom\0906-RTX30\_internal\CUDNN\Win10.0;D:\workroom\0906-RTX30\_internal\CUDA;D:\workroom\0906-RTX30\_internal\CUDNN;D:\workroom\0906-RTX30\_internal\python-3.7;D:\workroom\0906-RTX30\_internal\python-3.7\Scripts;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.5\libnvvp;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Windows\System32\OpenSSH\;F:\Users\Win-10\anaconda3;F:\Program Files\Git\cmd;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\NVIDIA Corporation\NVIDIA NvDLISR;C:\Program Files\NVIDIA Corporation\Nsight Compute 2021.3.1\;C:\Users\Win-10\AppData\Local\Microsoft\WindowsApps;C:\Program Files\Multipass\bin

D:\workroom\0906-RTX30>SET WORKSPACE=D:\workroom\0906-RTX30\_internal\..\workspace

D:\workroom\0906-RTX30>SET DFL_ROOT=D:\workroom\0906-RTX30\_internal\DeepFaceLab

D:\workroom\0906-RTX30>"D:\workroom\0906-RTX30\_internal\python-3.7\python.exe" "D:\workroom\0906-RTX30\_internal\DeepFaceLab\main.py" train     --training-data-src-dir "D:\workroom\0906-RTX30\_internal\..\workspace\data_src\aligned"     --training-data-dst-dir "D:\workroom\0906-RTX30\_internal\..\workspace\data_dst\aligned"     --pretraining-data-dir "D:\workroom\0906-RTX30\_internal\pretrain_faces"     --model-dir "D:\workroom\0906-RTX30\_internal\..\workspace\model"     --model SAEHD
Running trainer.

Choose one of saved models, or enter a name to create a new model.
[r] : rename
[d] : delete

[0] : RTX3090 - latest
[1] : DF-UD - 320RES - Regular Trained - Whole Face - 1060k
: 0
0
Loading RTX3090_SAEHD model...

Choose one or several GPU idxs (separated by comma).

[CPU] : CPU
  [0] : NVIDIA GeForce RTX 3060

[0] Which GPU indexes to choose? :
0

Press enter in 2 seconds to override model settings.
[0] Autobackup every N hour ( 0..24 ?:help ) :
0
[y] Write preview history ( y/n ?:help ) : n
[0] Target iteration :
0
[n] Flip SRC faces randomly ( y/n ?:help ) :
n
[y] Flip DST faces randomly ( y/n ?:help ) :
y
[4] Batch_size ( ?:help ) :
4
[y] Masked training ( y/n ?:help ) :
y
[y] Eyes and mouth priority ( y/n ?:help ) :
y
[n] Uniform yaw distribution of samples ( y/n ?:help ) : y
[y] Blur out mask ( y/n ?:help ) : n
[y] Place models and optimizer on GPU ( y/n ?:help ) :
y
[y] Use AdaBelief optimizer? ( y/n ?:help ) :
y
[y] Use learning rate dropout ( n/y/cpu ?:help ) :
y
[n] Enable random warp of samples ( y/n ?:help ) : y
[0.0] Random hue/saturation/light intensity ( 0.0 .. 0.3 ?:help ) :
0.0
[0.1] GAN power ( 0.0 .. 5.0 ?:help ) :
0.1
[40] GAN patch size ( 3-640 ?:help ) :
40
[16] GAN dimensions ( 4-512 ?:help ) :
16
[0.0] Face style power ( 0.0..100.0 ?:help ) :
0.0
[0.0] Background style power ( 0.0..100.0 ?:help ) :
0.0
[lct] Color transfer for src faceset ( none/rct/lct/mkl/idt/sot ?:help ) :
lct
[n] Enable gradient clipping ( y/n ?:help ) :
n
[n] Enable pretraining mode ( y/n ?:help ) :
n
Initializing models: 100%|###############################################################| 7/7 [00:06<00:00,  1.01it/s]
Loading samples: 100%|############################################################| 1036/1036 [00:02<00:00, 407.70it/s]
Sort by yaw: 100%|#################################################################| 128/128 [00:00<00:00, 5112.23it/s]
Loading samples: 100%|##############################################################| 886/886 [00:02<00:00, 410.74it/s]
Sort by yaw: 100%|#################################################################| 128/128 [00:00<00:00, 5565.04it/s]
================== Model Summary ===================
==                                                ==
==            Model name: RTX3090_SAEHD           ==
==                                                ==
==     Current iteration: 4641219                 ==
==                                                ==
==---------------- Model Options -----------------==
==                                                ==
==            resolution: 320                     ==
==             face_type: wf                      ==
==     models_opt_on_gpu: True                    ==
==                 archi: liae-ud                 ==
==               ae_dims: 256                     ==
==                e_dims: 64                      ==
==                d_dims: 64                      ==
==           d_mask_dims: 22                      ==
==       masked_training: True                    ==
==       eyes_mouth_prio: True                    ==
==           uniform_yaw: True                    ==
==         blur_out_mask: False                   ==
==             adabelief: True                    ==
==            lr_dropout: y                       ==
==           random_warp: True                    ==
==       true_face_power: 0.0                     ==
==      face_style_power: 0.0                     ==
==        bg_style_power: 0.0                     ==
==               ct_mode: lct                     ==
==              clipgrad: False                   ==
==              pretrain: False                   ==
==       autobackup_hour: 0                       ==
== write_preview_history: False                   ==
==           target_iter: 0                       ==
==       random_src_flip: False                   ==
==       random_dst_flip: True                    ==
==            batch_size: 4                       ==
==             gan_power: 0.1                     ==
==        gan_patch_size: 40                      ==
==              gan_dims: 16                      ==
==      random_hsv_power: 0.0                     ==
==                                                ==
==------------------ Running On ------------------==
==                                                ==
==          Device index: 0                       ==
==                  Name: NVIDIA GeForce RTX 3060 ==
==                  VRAM: 10.24GB                 ==
==                                                ==
====================================================
Starting. Press "Enter" to stop training and save model.
[

新手剪辑师

搂住。你的模型是liae-ud 结构的。
liae更偏重于光影。
DF结构的更偏重于还原度。

ture face 那个参数，只支持DF结构。

我也是新手，希望能帮到你。

zhaitian520

原来是这样，get到了。

easonliu0203

學習學習

skwb2012

學習學習

wcw888

学到了，感谢分享。好像听说要开gan才有

1234qwer111

深度学习

xiaomei

这么高深莫测

07095

学到了

Bobsimth

谢谢解答