安裝kohya_SS gui

1.安裝python：https://www.python.org/ftp/python/3.10.9/python-3.10.9-amd64.exe

安裝git：https://git-scm.com/download/win

安裝：Visual Studio 2015, 2017, 2019, and 2022?redistributable

網(wǎng)址：https://aka.ms/vs/17/release/vc_redist.x64.exe

2.管理員打開(kāi)powershell,輸入Set-ExecutionPolicy Unrestricted，然后回車(chē)輸入Y

3.再次打開(kāi)安裝盤(pán)符,輸入例如cd D：（可以是任何盤(pán)符，存儲(chǔ)空間要大），切換到安裝盤(pán)符

4.創(chuàng)建文件夾例如D:\kohya_SS,

5.輸入D:\cd \kohya_SS切換到該文件夾，

6.然后復(fù)制git clone https://github.com/bmaltais/kohya_ss.git

cd kohya_ss

.\setup.bat

安裝

如果國(guó)內(nèi)網(wǎng)速限制請(qǐng)?jiān)贑：user/xx/創(chuàng)建pip環(huán)境

新建pip.txt把后綴改為.in然后用記事本打開(kāi)

粘貼代碼：

?

[global]

?

timeout = 6000

?

index-url = https://pypi.tuna.tsinghua.edu.cn/simple

?

trusted-host = pypi.tuna.tsinghua.edu.cn

就可以了

下面就是運(yùn)行第5步的setup.bat

會(huì)出現(xiàn)

1. This machine

? ? AWS (Amazon SageMaker)

點(diǎn)擊回車(chē)

2. No distributed training

? ? multi-CPU

? ? multi-GPU

? ? TPU

多張顯卡就根據(jù)情況選擇，沒(méi)有就按下回車(chē)

3.Do you want to run your training on CPU only（你只使用GPU訓(xùn)練嗎） (even if a GPU / Apple Silicon device is available)? [yes/NO]:no? ?

4.Do you wish to optimize your script with torch dynamo?[yes/NO]:no? ??

5.Do you want to use DeepSpeed? [yes/NO]: no

6.What GPU(s) (by id) should be used for training on this machine as a comma-seperated list? [all]: all

7.Please select a choice using the arrow or number keys, and selecting with enter,no

?* no

? ? fp16

? ? bf16

? ? fp8

cudnn安裝30/40系顯卡配置，https://b1.thefileditch.ch/mwxKTEtelILoIbMbruuM.zip下載放到kohya_SS文件夾下

8.然后C:\Users\Administrator\kohya_ss>

.\venv\Scripts\activate

python .\tools\cudann_1.8_install.py

粘貼代碼安裝這樣就安裝完成啦，torch2.0自帶cudnn無(wú)需安裝。

9.然后運(yùn)行g(shù)ui-user.bat就打開(kāi)UI界面啦。