一、安装配置Anaconda
进入官网下载安装包https://www.anaconda.com/并安装，然后将Anaconda配置到环境变量中。
打开命令行，依次通过如下命令创建Python运行虚拟环境。
conda env create novelai python==3.10.6
1
E:\workspace\02_Python\novalai>conda info –envs
# conda environments:
#
base * D:\anaconda3
novelai D:\anaconda3\envs\novelai
1
2
3
4
5
conda activate novelai
1
二、安装CUDA
笔者的显卡为NVIDIA，需安装NVIDIA的开发者工具进入官网https://developer.nvidia.com/，根据自己计算机的系统情况，选择合适的安装包下载安装。
打开安装程序后，依照提示完成安装。
安装完成后，在命令窗口输入如下命令，输出CUDA版本即安装成功。
C:\Users\yefuf>nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Wed_Sep_21_10:41:10_Pacific_Daylight_Time_2022
Cuda compilation tools, release 11.8, V11.8.89
Build cuda_11.8.r11.8/compiler.31833905_0
1
2
3
4
5
6
三、安装pytorch
进入官网https://pytorch.org/，根据计算机配置选择合适的版本进行安装。这里需要注意的是CUDA的平台选择，先打开NVIDIA控制面板-帮助-系统信息-组件查看CUDA版本，官网上选择的计算平台需要低于计算机的NVIDIA版本。
配置选择完成后，官网会生成相应的安装命令。
将安装命令复制出，命令窗口执行安装即可。
conda install pytorch torchvision torchaudio pytorch-cuda=11.6 -c pytorch -c nvidia
1
当查到Pytorch官网推荐的CUDA版本跟你的显卡版本不匹配时，就需要根据官网的CUDA版本找到对应的显卡驱动版本并升级显卡驱动，对应关系可通过https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html查看
四、安装git
进入git官网https://git-scm.com/，下载安装即可。
五、搭建stable-diffusion-webui
进入项目地址https://github.com/AUTOMATIC1111/stable-diffusion-webui，通过git将项目克隆下来。
git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
Cloning into ‘stable-diffusion-webui’…
remote: Enumerating objects: 10475, done.
remote: Counting objects: 100% (299/299), done.
remote: Compressing objects: 100% (199/199), done.
remote: Total 10475 (delta 178), reused 199 (delta 100), pack-reused 10176
Receiving objects: 100% (10475/10475), 23.48 MiB | 195.00 KiB/s, done.
Resolving deltas: 100% (7312/7312), done.
1
2
3
4
5
6
7
8
克隆下载扩展库。
git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui-aesthetic-gradients “extensions/aesthetic-gradients”
Cloning into ‘extensions/aesthetic-gradients’…
remote: Enumerating objects: 21, done.
remote: Counting objects: 100% (21/21), done.
remote: Compressing objects: 100% (12/12), done.
remote: Total 21 (delta 3), reused 18 (delta 3), pack-reused 0
Receiving objects: 100% (21/21), 1.09 MiB | 1.34 MiB/s, done.
Resolving deltas: 100% (3/3), done.
1
2
3
4
5
6
7
8
git clone https://github.com/yfszzx/stable-diffusion-webui-images-browser “extensions/images-browser”
Cloning into ‘extensions/images-browser’…
remote: Enumerating objects: 118, done.
remote: Counting objects: 100% (118/118), done.
remote: Compressing objects: 100% (70/70), done.
remote: Total 118 (delta 42), reused 65 (delta 24), pack-reused 0
Receiving objects: 100% (118/118), 33.01 KiB | 476.00 KiB/s, done.
Resolving deltas: 100% (42/42), done.
1
2
3
4
5
6
7
8
克隆完成后，extensions目录会多如下文件夹：
下载模型库https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Dependencies，并将下载的.ckpt
放到models/Stable-diffusion文件夹中。模型很大，推荐使用下载器。
安装项目所需的Python依赖库。
pip install -r requirements.txt
1
安装完成之后，运行如下命令，顺利的话，当程序加载完成模型之后，会自动打开http://127.0.0.1:7860/显示平台主页。
python launch.py –autolaunch
1
进入平台的设置页面，选择语言为中文，重启程序之后，即可看到页面显示为中文。
在界面中输入作画内容的正向提示词（画想要什么特征）和反向提示词（画不想要什么特征），点击生成即可开始自动作画。
如上述的提示词作出的画如图（由于随机种子不同，生成的画会有差异）。
六、如何设置提示词
这里建议使用元素法典https://docs.qq.com/doc/DWHl3am5Zb05QbGVs，上面有前人整理好的提示词及效果，以供参考。
七、可能遇到的问题
1、GitHub访问不了或访问慢
一般为DNS解析问题，需要修改本地host文件，增加配置内容，绕过域名解析，达到加速访问的目的。
访问https://www.ipaddress.com/，分别输入github.com和github.global.ssl.fastly.net，获取域名对应的IP地址。
打开系统的Host文件，将IP和域名的对应关系配置到Host文件中。
配置文件内容如下：
140.82.114.4 github.com
199.232.5.194 github.global.ssl.fastly.net
1
2
执行命令ipconfig /flushdns刷新DNS即可。
2、pip安装依赖库慢或常下载失败
pip安装依赖库时默认选择国外的源，安装速度会非常慢，可以考虑切换为国内源，常用的国内源如下：
阿里云 https://mirrors.aliyun.com/pypi/simple/
中国科技大学 https://pypi.mirrors.ustc.edu.cn/simple/
豆瓣(douban) https://pypi.douban.com/simple/
清华大学 https://pypi.tuna.tsinghua.edu.cn/simple/
中国科学技术大学 https://pypi.mirrors.ustc.edu.cn/simple/
1
2
3
4
5
在安装依赖库时，可使用pip install -i 源 空格 安装包名称进行源的选择，如pip install -i https://mirrors.aliyun.com/pypi/simple numpy。
也可以通过增加配置文件，使安装依赖库时默认选择国内的源，在用户目录下增加pip.ini文件。
在文件中写入如下内容。
[global]
timeout = 60000
index-url = https://pypi.tuna.tsinghua.edu.cn/simple
[install]
use-mirrors = true
mirrors = https://pypi.tuna.tsinghua.edu.cn
1
2
3
4
5
6
7
3、安装CLIP时提示Connection was aborted, errno 10053
出错时的错误打印如下：
(novelai) E:\workspace\02_Python\novalai\stable-diffusion-webui>python launch.py
Python 3.10.6 | packaged by conda-forge | (main, Oct 24 2022, 16:02:16) [MSC v.1916 64 bit (AMD64)]
Commit hash: b8f2dfed3c0085f1df359b9dc5b3841ddc2196f0
Installing clip
Traceback (most recent call last):
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 251, in <module>
prepare_enviroment()
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 178, in prepare_enviroment
run_pip(f”install {clip_package}”, “clip”)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 63, in run_pip
return run(f'”{python}” -m pip {args} –prefer-binary{index_url_line}’, desc=f”Installing {desc}”, errdesc=f”Couldn’t install {desc}”)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 34, in run
raise RuntimeError(message)
RuntimeError: Couldn’t install clip.
Command: “D:\anaconda3\envs\novelai\python.exe” -m pip install git+https://github.com/openai/CLIP.git@d50d76daa670286dd6cacf3bcd80b5e4823fc8e1 –prefer-binary
Error code: 1
stdout: Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
Collecting git+https://github.com/openai/CLIP.git@d50d76daa670286dd6cacf3bcd80b5e4823fc8e1
Cloning https://github.com/openai/CLIP.git (to revision d50d76daa670286dd6cacf3bcd80b5e4823fc8e1) to c:\users\yefuf\appdata\local\temp\pip-req-build-f8w7kbzg
stderr: Running command git clone –filter=blob:none –quiet https://github.com/openai/CLIP.git ‘C:\Users\yefuf\AppData\Local\Temp\pip-req-build-f8w7kbzg’
fatal: unable to access ‘https://github.com/openai/CLIP.git/’: OpenSSL SSL_read: Connection was aborted, errno 10053
error: subprocess-exited-with-error
git clone –filter=blob:none –quiet https://github.com/openai/CLIP.git ‘C:\Users\yefuf\AppData\Local\Temp\pip-req-build-f8w7kbzg’ did not run successfully.
exit code: 128
See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error
git clone –filter=blob:none –quiet https://github.com/openai/CLIP.git ‘C:\Users\yefuf\AppData\Local\Temp\pip-req-build-f8w7kbzg’ did not run successfully.
exit code: 128
See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
通过访CLIP项目GitHub主页，发现该项目可以通过如下命令进行安装解决。
pip install ftfy regex tqdm
pip install git+https://github.com/openai/CLIP.git
1
2
4、项目启动中提示Connection was reset in connection to github.com
出错时的错误打印如下：
(novelai) E:\workspace\02_Python\novalai\stable-diffusion-webui>python launch.py
Python 3.10.6 | packaged by conda-forge | (main, Oct 24 2022, 16:02:16) [MSC v.1916 64 bit (AMD64)]
Commit hash: b8f2dfed3c0085f1df359b9dc5b3841ddc2196f0
Cloning Stable Diffusion into repositories\stable-diffusion…
Cloning Taming Transformers into repositories\taming-transformers…
Traceback (most recent call last):
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 251, in <module>
prepare_enviroment()
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 201, in prepare_enviroment
git_clone(taming_transformers_repo, repo_dir(‘taming-transformers’), “Taming Transformers”, taming_transformers_commit_hash)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 85, in git_clone
run(f'”{git}” clone “{url}” “{dir}”‘, f”Cloning {name} into {dir}…”, f”Couldn’t clone {name}”)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 34, in run
raise RuntimeError(message)
RuntimeError: Couldn’t clone Taming Transformers.
Command: “git” clone “https://github.com/CompVis/taming-transformers.git” “repositories\taming-transformers”
Error code: 128
stdout: <empty>
stderr: Cloning into ‘repositories\taming-transformers’…
fatal: unable to access ‘https://github.com/CompVis/taming-transformers.git/’: OpenSSL SSL_connect: Connection was reset in connection to github.com:443
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
在命令窗口中输入如下命令，然后重新运行程序，但实际操作下来，仍有较大概率在克隆项目的过程中失败。
git config –global http.postBuffer 524288000
git config –global http.sslVerify “false”
1
2
查看lauch.py中的代码可以发现，程序在启动时有对依赖项目进行检查，如项目不存在，则克隆下来。
def prepare_enviroment():
torch_command = os.environ.get(‘TORCH_COMMAND’, “pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 –extra-index-url https://download.pytorch.org/whl/cu113”)
requirements_file = os.environ.get(‘REQS_FILE’, “requirements_versions.txt”)
commandline_args = os.environ.get(‘COMMANDLINE_ARGS’, “”)
gfpgan_package = os.environ.get(‘GFPGAN_PACKAGE’, “git+https://github.com/TencentARC/GFPGAN.git@8d2447a2d918f8eba5a4a01463fd48e45126a379”)
clip_package = os.environ.get(‘CLIP_PACKAGE’, “git+https://github.com/openai/CLIP.git@d50d76daa670286dd6cacf3bcd80b5e4823fc8e1”)
deepdanbooru_package = os.environ.get(‘DEEPDANBOORU_PACKAGE’, “git+https://github.com/KichangKim/DeepDanbooru.git@d91a2963bf87c6a770d74894667e9ffa9f6de7ff”)
xformers_windows_package = os.environ.get(‘XFORMERS_WINDOWS_PACKAGE’, ‘https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/f/xformers-0.0.14.dev0-cp310-cp310-win_amd64.whl’)
stable_diffusion_repo = os.environ.get(‘STABLE_DIFFUSION_REPO’, “https://github.com/CompVis/stable-diffusion.git”)
taming_transformers_repo = os.environ.get(‘TAMING_REANSFORMERS_REPO’, “https://github.com/CompVis/taming-transformers.git”)
k_diffusion_repo = os.environ.get(‘K_DIFFUSION_REPO’, ‘https://github.com/crowsonkb/k-diffusion.git’)
codeformer_repo = os.environ.get(‘CODEFORMET_REPO’, ‘https://github.com/sczhou/CodeFormer.git’)
blip_repo = os.environ.get(‘BLIP_REPO’, ‘https://github.com/salesforce/BLIP.git’)
stable_diffusion_commit_hash = os.environ.get(‘STABLE_DIFFUSION_COMMIT_HASH’, “69ae4b35e0a0f6ee1af8bb9a5d0016ccb27e36dc”)
taming_transformers_commit_hash = os.environ.get(‘TAMING_TRANSFORMERS_COMMIT_HASH’, “24268930bf1dce879235a7fddd0b2355b84d7ea6”)
k_diffusion_commit_hash = os.environ.get(‘K_DIFFUSION_COMMIT_HASH’, “f4e99857772fc3a126ba886aadf795a332774878”)
codeformer_commit_hash = os.environ.get(‘CODEFORMER_COMMIT_HASH’, “c5b4593074ba6214284d6acd5f1719b6c5d739af”)
blip_commit_hash = os.environ.get(‘BLIP_COMMIT_HASH’, “48211a1594f1321b00f14c9f7a5b4813144b2fb9”)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
因此，我们打开git bash重新执行上述两条git命令，预先将项目克隆下来。
git clone https://github.com/CompVis/taming-transformers.git “repositories\taming-transformers”
git clone https://github.com/crowsonkb/k-diffusion.git “repositories\k-diffusion”
git clone https://github.com/sczhou/CodeFormer.git “repositories\CodeFormer”
git clone https://github.com/salesforce/BLIP.git “repositories\BLIP”
1
2
3
4
5
6
7
克隆完成之后如图：
5、项目启动中提示CUDA out of memory
出错时的错误打印如下：
(novelai) E:\workspace\02_Python\novalai\stable-diffusion-webui>python launch.py
Python 3.10.6 | packaged by conda-forge | (main, Oct 24 2022, 16:02:16) [MSC v.1916 64 bit (AMD64)]
Commit hash: b8f2dfed3c0085f1df359b9dc5b3841ddc2196f0
Fetching updates for BLIP…
Checking out commit for BLIP with hash: 48211a1594f1321b00f14c9f7a5b4813144b2fb9…
Installing requirements for CodeFormer
Installing requirements for Web UI
Launching Web UI with arguments:
Moving sd-v1-4.ckpt from E:\workspace\02_Python\novalai\stable-diffusion-webui\models to E:\workspace\02_Python\novalai\stable-diffusion-webui\models\Stable-diffusion.
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
making attention of type ‘vanilla’ with 512 in_channels
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type ‘vanilla’ with 512 in_channels
Downloading: 100%|██████████████████████████████████████████████████████████████████| 939k/939k [00:00<00:00, 1.26MB/s]
Downloading: 100%|███████████████████████████████████████████████████████████████████| 512k/512k [00:01<00:00, 344kB/s]
Downloading: 100%|████████████████████████████████████████████████████████████████████████████| 389/389 [00:00<?, ?B/s]
Downloading: 100%|████████████████████████████████████████████████████████████████████████████| 905/905 [00:00<?, ?B/s]
Downloading: 100%|████████████████████████████████████████████████████████████████████████| 4.41k/4.41k [00:00<?, ?B/s]
Downloading: 100%|████████████████████████████████████████████████████████████████| 1.59G/1.59G [03:56<00:00, 7.23MB/s]
Loading weights [7460a6fa] from E:\workspace\02_Python\novalai\stable-diffusion-webui\models\Stable-diffusion\sd-v1-4.ckpt
Global Step: 470000
Traceback (most recent call last):
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 252, in <module>
start()
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\launch.py”, line 247, in start
webui.webui()
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\webui.py”, line 148, in webui
initialize()
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\webui.py”, line 83, in initialize
modules.sd_models.load_model()
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\sd_models.py”, line 252, in load_model
sd_model.to(shared.device)
File “D:\anaconda3\envs\novelai\lib\site-packages\pytorch_lightning\core\mixins\device_dtype_mixin.py”, line 113, in to
return super().to(*args, **kwargs)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 987, in to
return self._apply(convert)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 639, in _apply
module._apply(fn)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 639, in _apply
module._apply(fn)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 639, in _apply
module._apply(fn)
[Previous line repeated 2 more times]
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 662, in _apply
param_applied = fn(param)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 985, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 2.00 GiB total capacity; 1.68 GiB already allocated; 0 bytes free; 1.72 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
根据提示，先尝试用如下命令改变pytorch配置，仍旧报错!
set PYTORCH_CUDA_ALLOC_CONF=garbage_collection_threshold:0.6,max_split_size_mb:128
1
尝试增加代码with torch.no_grad()，使内存就不会分配参数梯度的空间，仍旧报错！
由于提示内存溢出，先通过控制面板->所有控制面板项->管理工具->系统信息，查看显卡内存大小。
官方推荐的显卡内存大小为4GB以上，而笔者的显卡内存只有2GB，显然GPU不符合要求。查看项目的命令选项，发现项目支持CPU计算–use-cpu。
(novelai) E:\workspace\02_Python\novalai\stable-diffusion-webui>python launch.py -h
Python 3.10.6 | packaged by conda-forge | (main, Oct 24 2022, 16:02:16) [MSC v.1916 64 bit (AMD64)]
Commit hash: b8f2dfed3c0085f1df359b9dc5b3841ddc2196f0
Installing requirements for Web UI
Launching Web UI with arguments: -h
usage: launch.py [-h] [–config CONFIG] [–ckpt CKPT] [–ckpt-dir CKPT_DIR] [–gfpgan-dir GFPGAN_DIR]
[–gfpgan-model GFPGAN_MODEL] [–no-half] [–no-half-vae] [–no-progressbar-hiding]
[–max-batch-count MAX_BATCH_COUNT] [–embeddings-dir EMBEDDINGS_DIR]
[–hypernetwork-dir HYPERNETWORK_DIR] [–localizations-dir LOCALIZATIONS_DIR] [–allow-code]
[–medvram] [–lowvram] [–lowram] [–always-batch-cond-uncond] [–unload-gfpgan]
[–precision {full,autocast}] [–share] [–ngrok NGROK] [–ngrok-region NGROK_REGION]
[–enable-insecure-extension-access] [–codeformer-models-path CODEFORMER_MODELS_PATH]
[–gfpgan-models-path GFPGAN_MODELS_PATH] [–esrgan-models-path ESRGAN_MODELS_PATH]
[–bsrgan-models-path BSRGAN_MODELS_PATH] [–realesrgan-models-path REALESRGAN_MODELS_PATH]
[–scunet-models-path SCUNET_MODELS_PATH] [–swinir-models-path SWINIR_MODELS_PATH]
[–ldsr-models-path LDSR_MODELS_PATH] [–clip-models-path CLIP_MODELS_PATH] [–xformers]
[–force-enable-xformers] [–deepdanbooru] [–opt-split-attention] [–opt-split-attention-invokeai]
[–opt-split-attention-v1] [–disable-opt-split-attention]
[–use-cpu {all,sd,interrogate,gfpgan,swinir,esrgan,scunet,codeformer} [{all,sd,interrogate,gfpgan,swinir,esrgan,scunet,codeformer} …]]
[–listen] [–port PORT] [–show-negative-prompt] [–ui-config-file UI_CONFIG_FILE]
[–hide-ui-dir-config] [–freeze-settings] [–ui-settings-file UI_SETTINGS_FILE] [–gradio-debug]
[–gradio-auth GRADIO_AUTH] [–gradio-img2img-tool {color-sketch,editor}] [–opt-channelslast]
[–styles-file STYLES_FILE] [–autolaunch] [–theme THEME] [–use-textbox-seed]
[–disable-console-progressbars] [–enable-console-prompts] [–vae-path VAE_PATH]
[–disable-safe-unpickle] [–api] [–nowebui] [–ui-debug-mode] [–device-id DEVICE_ID]
[–administrator] [–cors-allow-origins CORS_ALLOW_ORIGINS] [–tls-keyfile TLS_KEYFILE]
[–tls-certfile TLS_CERTFILE] [–server-name SERVER_NAME]
options:
-h, –help show this help message and exit
–config CONFIG path to config which constructs model
–ckpt CKPT path to checkpoint of stable diffusion model; if specified, this checkpoint will be added to
the list of checkpoints and loaded
–ckpt-dir CKPT_DIR Path to directory with stable diffusion checkpoints
–gfpgan-dir GFPGAN_DIR
GFPGAN directory
–gfpgan-model GFPGAN_MODEL
GFPGAN model file name
–no-half do not switch the model to 16-bit floats
–no-half-vae do not switch the VAE model to 16-bit floats
–no-progressbar-hiding
do not hide progressbar in gradio UI (we hide it because it slows down ML if you have hardware
acceleration in browser)
–max-batch-count MAX_BATCH_COUNT
maximum batch count value for the UI
–embeddings-dir EMBEDDINGS_DIR
embeddings directory for textual inversion (default: embeddings)
–hypernetwork-dir HYPERNETWORK_DIR
hypernetwork directory
–localizations-dir LOCALIZATIONS_DIR
localizations directory
–allow-code allow custom script execution from webui
–medvram enable stable diffusion model optimizations for sacrificing a little speed for low VRM usage
–lowvram enable stable diffusion model optimizations for sacrificing a lot of speed for very low VRM
usage
–lowram load stable diffusion checkpoint weights to VRAM instead of RAM
–always-batch-cond-uncond
disables cond/uncond batching that is enabled to save memory with –medvram or –lowvram
–unload-gfpgan does not do anything.
–precision {full,autocast}
evaluate at this precision
–share use share=True for gradio and make the UI accessible through their site
–ngrok NGROK ngrok authtoken, alternative to gradio –share
–ngrok-region NGROK_REGION
The region in which ngrok should start.
–enable-insecure-extension-access
enable extensions tab regardless of other options
–codeformer-models-path CODEFORMER_MODELS_PATH
Path to directory with codeformer model file(s).
–gfpgan-models-path GFPGAN_MODELS_PATH
Path to directory with GFPGAN model file(s).
–esrgan-models-path ESRGAN_MODELS_PATH
Path to directory with ESRGAN model file(s).
–bsrgan-models-path BSRGAN_MODELS_PATH
Path to directory with BSRGAN model file(s).
–realesrgan-models-path REALESRGAN_MODELS_PATH
Path to directory with RealESRGAN model file(s).
–scunet-models-path SCUNET_MODELS_PATH
Path to directory with ScuNET model file(s).
–swinir-models-path SWINIR_MODELS_PATH
Path to directory with SwinIR model file(s).
–ldsr-models-path LDSR_MODELS_PATH
Path to directory with LDSR model file(s).
–clip-models-path CLIP_MODELS_PATH
Path to directory with CLIP model file(s).
–xformers enable xformers for cross attention layers
–force-enable-xformers
enable xformers for cross attention layers regardless of whether the checking code thinks you
can run it; do not make bug reports if this fails to work
–deepdanbooru enable deepdanbooru interrogator
–opt-split-attention
force-enables Doggettx’s cross-attention layer optimization. By default, it’s on for torch
cuda.
–opt-split-attention-invokeai
force-enables InvokeAI’s cross-attention layer optimization. By default, it’s on when cuda is
unavailable.
–opt-split-attention-v1
enable older version of split attention optimization that does not consume all the VRAM it can
find
–disable-opt-split-attention
force-disables cross-attention layer optimization
–use-cpu {all,sd,interrogate,gfpgan,swinir,esrgan,scunet,codeformer} [{all,sd,interrogate,gfpgan,swinir,esrgan,scunet,codeformer} …]
use CPU as torch device for specified modules
–listen launch gradio with 0.0.0.0 as server name, allowing to respond to network requests
–port PORT launch gradio with given server port, you need root/admin rights for ports < 1024, defaults to
7860 if available
–show-negative-prompt
does not do anything
–ui-config-file UI_CONFIG_FILE
filename to use for ui configuration
–hide-ui-dir-config hide directory configuration from webui
–freeze-settings disable editing settings
–ui-settings-file UI_SETTINGS_FILE
filename to use for ui settings
–gradio-debug launch gradio with –debug option
–gradio-auth GRADIO_AUTH
set gradio authentication like “username:password”; or comma-delimit multiple like
“u1:p1,u2:p2,u3:p3”
–gradio-img2img-tool {color-sketch,editor}
gradio image uploader tool: can be either editor for ctopping, or color-sketch for drawing
–opt-channelslast change memory type for stable diffusion to channels last
–styles-file STYLES_FILE
filename to use for styles
–autolaunch open the webui URL in the system’s default browser upon launch
–theme THEME launches the UI with light or dark theme
–use-textbox-seed use textbox for seeds in UI (no up/down, but possible to input long seeds)
–disable-console-progressbars
do not output progressbars to console
–enable-console-prompts
print prompts to console when generating with txt2img and img2img
–vae-path VAE_PATH Path to Variational Autoencoders model
–disable-safe-unpickle
disable checking pytorch models for malicious code
–api use api=True to launch the api with the webui
–nowebui use api=True to launch the api instead of the webui
–ui-debug-mode Don’t load model to quickly launch UI
–device-id DEVICE_ID
Select the default CUDA device to use (export CUDA_VISIBLE_DEVICES=0,1,etc might be needed
before)
–administrator Administrator rights
–cors-allow-origins CORS_ALLOW_ORIGINS
Allowed CORS origins
–tls-keyfile TLS_KEYFILE
Partially enables TLS, requires –tls-certfile to fully function
–tls-certfile TLS_CERTFILE
Partially enables TLS, requires –tls-keyfile to fully function
–server-name SERVER_NAME
Sets hostname of server
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
尝试构造如下运行参数，–use-cpu all使所有模块均使用CPU计算，–lowram –always-batch-cond-uncond使用低内存配置选项，程序可以成功运行。
(novelai) E:\workspace\02_Python\novalai\stable-diffusion-webui>python launch.py –lowram –always-batch-cond-uncond –use-cpu all
Python 3.10.6 | packaged by conda-forge | (main, Oct 24 2022, 16:02:16) [MSC v.1916 64 bit (AMD64)]
Commit hash: b8f2dfed3c0085f1df359b9dc5b3841ddc2196f0
Installing requirements for Web UI
Launching Web UI with arguments: –lowram –always-batch-cond-uncond –use-cpu all
Warning: caught exception ‘Expected a cuda device, but got: cpu’, memory monitor disabled
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
making attention of type ‘vanilla’ with 512 in_channels
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type ‘vanilla’ with 512 in_channels
Loading weights [7460a6fa] from E:\workspace\02_Python\novalai\stable-diffusion-webui\models\Stable-diffusion\sd-v1-4.ckpt
Global Step: 470000
Applying cross attention optimization (Doggettx).
Model loaded.
Loaded a total of 0 textual inversion embeddings.
Embeddings:
Running on local URL: http://127.0.0.1:7860
To create a public link, set `share=True` in `launch()`.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
然而，开始作画时提示RuntimeError: “LayerNormKernelImpl” not implemented for ‘Half’错误！如果安装网上的处理方法，将half函数在工程中替换为float函数，则会出现device不匹配问题。
Traceback (most recent call last):
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\ui.py”, line 185, in f
res = list(func(*args, **kwargs))
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\webui.py”, line 57, in f
res = func(*args, **kwargs)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\txt2img.py”, line 48, in txt2img
processed = process_images(p)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\processing.py”, line 423, in process_images
res = process_images_inner(p)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\processing.py”, line 508, in process_images_inner
uc = prompt_parser.get_learned_conditioning(shared.sd_model, len(prompts) * [p.negative_prompt], p.steps)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\prompt_parser.py”, line 138, in get_learned_conditioning
conds = model.get_learned_conditioning(texts)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\repositories\stable-diffusion\ldm\models\diffusion\ddpm.py”, line 558, in get_learned_conditioning
c = self.cond_stage_model(c)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 1190, in _call_impl
return forward_call(*input, **kwargs)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\sd_hijack.py”, line 338, in forward
z1 = self.process_tokens(tokens, multipliers)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\extensions\aesthetic-gradients\aesthetic_clip.py”, line 202, in __call__
z = self.process_tokens(remade_batch_tokens, multipliers)
File “E:\workspace\02_Python\novalai\stable-diffusion-webui\modules\sd_hijack.py”, line 353, in process_tokens
outputs = self.wrapped.transformer(input_ids=tokens, output_hidden_states=-opts.CLIP_stop_at_last_layers)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 1190, in _call_impl
return forward_call(*input, **kwargs)
File “D:\anaconda3\envs\novelai\lib\site-packages\transformers\models\clip\modeling_clip.py”, line 722, in forward
return self.text_model(
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 1190, in _call_impl
return forward_call(*input, **kwargs)
File “D:\anaconda3\envs\novelai\lib\site-packages\transformers\models\clip\modeling_clip.py”, line 643, in forward
encoder_outputs = self.encoder(
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 1190, in _call_impl
return forward_call(*input, **kwargs)
File “D:\anaconda3\envs\novelai\lib\site-packages\transformers\models\clip\modeling_clip.py”, line 574, in forward
layer_outputs = encoder_layer(
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 1190, in _call_impl
return forward_call(*input, **kwargs)
File “D:\anaconda3\envs\novelai\lib\site-packages\transformers\models\clip\modeling_clip.py”, line 316, in forward
hidden_states = self.layer_norm1(hidden_states)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\module.py”, line 1190, in _call_impl
return forward_call(*input, **kwargs)
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\modules\normalization.py”, line 190, in forward
return F.layer_norm(
File “D:\anaconda3\envs\novelai\lib\site-packages\torch\nn\functional.py”, line 2515, in layer_norm
return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
RuntimeError: “LayerNormKernelImpl” not implemented for ‘Half’
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
考虑到–use-cpu参数可以指定模块，则尝试使工程中的部分模块用CPU计算，其余在可用内存方位内用GPU计算，最终构造参数如下，项目可成功作画。
然而，此方式作画效率非常低，一般每张图片约5-6分钟。当参数设置较大时，会达到数小时。因此如果有条件可以升级计算机的显卡配置，或租赁云服务器效果会更好。
(novelai) E:\workspace\02_Python\novalai\stable-diffusion-webui>python launch.py –lowram –always-batch-cond-uncond –precision full –no-half –opt-split-attention-v1 –use-cpu sd –autolaunch
Python 3.10.6 | packaged by conda-forge | (main, Oct 24 2022, 16:02:16) [MSC v.1916 64 bit (AMD64)]
Commit hash: b8f2dfed3c0085f1df359b9dc5b3841ddc2196f0
Installing requirements for Web UI
Launching Web UI with arguments: –lowram –always-batch-cond-uncond –precision full –no-half –opt-split-attention-v1 –use-cpu sd
Warning: caught exception ‘Expected a cuda device, but got: cpu’, memory monitor disabled
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
making attention of type ‘vanilla’ with 512 in_channels
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type ‘vanilla’ with 512 in_channels
Loading weights [7460a6fa] from E:\workspace\02_Python\novalai\stable-diffusion-webui\models\Stable-diffusion\sd-v1-4.ckpt
Global Step: 470000
Applying v1 cross attention optimization.
Model loaded.
Loaded a total of 0 textual inversion embeddings.
Embeddings:
Running on local URL: http://127.0.0.1:7860
To create a public link, set `share=True` in `launch()`.
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [06:30<00:00, 19.50s/it]
Total progress: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [06:10<00:00, 18.51s/it]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
参考文献：
AI作画保姆级教程来了！逆天，太强了！
【作者：墨叶扶风http://blog.csdn.net/yefufeng】
————————————————
版权声明：本文为CSDN博主「yefufeng」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/yefufeng/article/details/127719952