![Board logo](images/green001/logo.png)
标题: [教程] 1. 新手进阶版入门教程:让AI小姐姐为你脱掉汉服,分开双腿(不需要显卡)[20P] [打印本页]
作者: 内力煮火锅 时间: 2023-3-24 12:27 标题: 1. 新手进阶版入门教程:让AI小姐姐为你脱掉汉服,分开双腿(不需要显卡)[20P]
[pp=请注意:]由于现在Colab开始对运行Stable Diffusion进行限制了,
所以教程中的登录Colab并运行安装指令这一步可能得要改成在自己电脑上安装Stable Diffusion了。
教程在这一步之后的内容还是有效的,自己电脑上安装Stable Diffusion之后继续按照教程操作即可。[/pp]
成品示例:
[attach]3937603[/attach]
为避免你在公共场所打开此贴尴尬,不穿衣服的作品在本教程靠后的位置。
本教程为【新手进阶版】。
学习本教程之前,请先学习完[bcolor=viewthread.php?tid=11590974&page=1#pid111725007,Magenta] 0. 新手零基础入门教程:如何不用显卡做出一张街头露出的AI图[13P][/bcolor]
开始教程前,先解答一些回复中问到的问题:
经过多次测试后发现,目前长和宽在800 X 800以内是安全的,可以适当增大其中一个数的同时减小另一个数。
引用:
看例图里手掌部分出现畸形的比较少,这是不是经过筛选后的?
筛选掉了不少风格相近的图,但是没有专门筛选掉不合格的图。
教程里的负面提示词,有很多关于手指还有肢体的,这些负面提示词可以尽最大可能地减少畸形的情况。
如果出图的时候发现哪些地方有畸形,就把那个地方的畸形描述进负面提示词里面,再下次出图的时候就能减少这种畸形。
引用:
教程出来的妹子脸长的都不一样,都是散图,怎样能让AI生成的都是同一个人?
按教程生成的图,下载下来以后,看一下图片的文件名,比方说文件名是00025-1903663517.png,那1903663517就是这张图的随机数种子。
你可以把这个数放进Seed那一格,其它的参数保持跟生成这张图的时候的参数一样(提示词可以微调),这样就有概率生成一个长相相似的人。
正文开始:
本次教程仍然使用跟上一次教程相同的方法,但是安装指令、提示词和反向提示词都改变了。
先登录https://colab.research.google.com/ (这一步有可能需要科学上网)
注意:如果你是操作完上一个教程后,直接来这个教程的,请先刷新一下Colab的网页,
让它重新连接,然后再按照上一个教程里的方法和这个教程里的安装指令一步一步操作下去。
详细说明和成品如下:
本次教程在安装指令中增加了Lora的模型,安装指令如下:复制内容到剪贴板
代码:
!pip install --upgrade fastapi==0.90.1
!git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui
!git clone https://github.com/yfszzx/stable-diffusion-webui-images-browser /content/stable-diffusion-webui/extensions/stable-diffusion-webui-images-browser
!curl -Lo chilloutmixni.safetensors https://huggingface.co/nolanaatama/chomni/resolve/main/chomni.safetensors
!curl -Lo ulzzang-6500.pt https://huggingface.co/nolanaatama/chomni/resolve/main/ulzzang-6500.pt
!curl -Lo koreanDollLikeness_v10.safetensors https://huggingface.co/duthanhduoc/chilloutmix-set/resolve/main/koreanDollLikeness_v10.safetensors
!curl -Lo koreanDollLikeness_v15.safetensors https://huggingface.co/amornlnw7/koreanDollLikeness_v15/resolve/main/koreanDollLikeness_v15.safetensors
!curl -Lo japaneseDollLikeness_v10.safetensors https://huggingface.co/aimainia/japaneseDollLikeness_v10/resolve/main/japaneseDollLikeness_v10.safetensors
!curl -Lo taiwanDollLikeness_v10.safetensors https://huggingface.co/jasonli007/taiwanDollLikeness_v10/resolve/main/taiwanDollLikeness_v10.safetensors
!mkdir /content/stable-diffusion-webui/models/Lora
!mv "/content/koreanDollLikeness_v10.safetensors" "/content/stable-diffusion-webui/models/Lora"
!mv "/content/koreanDollLikeness_v15.safetensors" "/content/stable-diffusion-webui/models/Lora"
!mv "/content/japaneseDollLikeness_v10.safetensors" "/content/stable-diffusion-webui/models/Lora"
!mv "/content/taiwanDollLikeness_v10.safetensors" "/content/stable-diffusion-webui/models/Lora"
!mv "/content/chilloutmixni.safetensors" "/content/stable-diffusion-webui/models/Stable-diffusion"
!mv "/content/ulzzang-6500.pt" "/content/stable-diffusion-webui/embeddings"
%cd /content/stable-diffusion-webui
!git checkout 91c8d0d
%pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchtext==0.14.1 torchaudio==0.13.1 torchdata==0.5.1 --extra-index-url https://download.pytorch.org/whl/cu117
!COMMANDLINE_ARGS="--share --disable-safe-unpickle --no-half-vae --xformers --reinstall-xformers --enable-insecure-extension-access" REQS_FILE="requirements.txt" python launch.py
Prompt(提示词)(上面的框框)(这里是让你填你想要的内容)(正常穿衣版,提示词增加了lora和Hanfu等):复制内容到剪贴板
代码:
<lora:japaneseDollLikeness_v10:0.3>,<lora:taiwanDollLikeness_v10:0.2>,<lora:koreanDollLikeness_v10:0.35>,<lora:koreanDollLikeness_v15:0.5>,best quality ,masterpiece, illustration, an extremely delicate and beautiful, extremely detailed ,CG ,unity ,8k wallpaper, Amazing, finely detail, masterpiece,best quality,official art,extremely detailed CG unity 8k wallpaper,absurdres, incredibly absurdres, huge filesize , ultra-detailed, highres, extremely detailed,beautiful detailed girl, extremely detailed eyes and face, beautiful detailed eyes,((smiling)),full-body shot,light on face,(Hanfu:1.1),1girl
注意:<lora:japaneseDollLikeness_v10:0.3>,<lora:taiwanDollLikeness_v10:0.2>,<lora:koreanDollLikeness_v10:0.35>,<lora:koreanDollLikeness_v15:0.5>这几个提示词是加载Lora模型用的,根据个人喜好,可以只挑一两个用,也可以全用。冒号后的数字是权重,可以根据个人喜好修改。
Negative prompt(反向提示词)(下面的框框)(这里是让你填你不想要的内容)(完善了上一个教程中的版本):复制内容到剪贴板
代码:
sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, bad anatomy,(long hair:1.4),DeepNegative,(fat:1.2),facing away, looking away,tilted head, {Multiple people}, lowres,bad anatomy,bad hands, text, error, missing fingers,extra digit, fewer digits, cropped, worstquality, low quality, normal quality,jpegartifacts,signature, watermark, username,blurry,bad feet,cropped,poorly drawn hands,poorly drawn face,mutation,deformed,worst quality,low quality,normal quality,jpeg artifacts,signature,watermark,extra fingers,fewer digits,extra limbs,extra arms,extra legs,malformed limbs,fused fingers,too many fingers,long neck,cross-eyed,mutated hands,polar lowres,bad body,bad proportions,gross proportions,text,error,missing fingers,missing arms,missing legs,extra digit, extra arms, extra leg, extra foot,((mole))
按照之前的教程和这次的安装指令和提示词,一步一步操作下来。
结果出的图如下:
[attach]3937604[/attach]
我觉得穿得太严实了而且室内有点沉闷了,于是增加了提示词(cleavage),outdoors
生成的作品见本帖第一张图。
我对这个作品比较满意,于是把这个图的种子随机数(1311099329)放进了Seed一栏,后面几张图都是用这个种子生成的:
[attach]3937605[/attach]
[attach]3937606[/attach]
我觉得妹子这么热的天你不如游个泳吧,于是保持这个Seed,提示词增加了((swimming))
[attach]3937607[/attach]
我觉得妹子你在泳池里可以穿得更少一些,于是提示词删掉了(Hanfu:1.1)
[attach]3937608[/attach]
我又想看看妹子穿汉服游泳,于是加回了(Hanfu:1.1)并删掉了(cleavage)
[attach]3937609[/attach]
妹子你笑得这么甜我的心脏要受不了啦,想看看妹子不笑的样子,删掉了((smiling))
[attach]3937610[/attach]
可以看出来,用相同seed生成的脸还是有点相像的。关于lora和Hanfu以及别的一些提示词大家也有了一定直观的感受。
下一个部分开始展示不穿衣服的汉服Lora。
不过在此之前,先给大家看一个辣眼睛的东西娱乐一下。
假如删掉Lora模型,seed改回随机,提示词里所有girl换成man,所有beautiful换成handsome会生成怎样的作品呢?
[attach]3937611[/attach]
可以看出来肉体变成男的了,但灵魂还是女的,并且还有一只手长反了,哈哈哈哈哈。
娱乐完了,下面开始不穿衣服的部分,顺便测试了一下生成不同尺寸的效果:
(Hanfu:1.1),outdoors改为((nude)),((spread legs))
可以看出来,用了Lora模型和一些相应的提示词以后,人物的脸比上一个教程少了很多混血感,变成了纯种的亚洲妹子。
[attach]3937612[/attach]
保持((nude)),((spread legs))并加回(Hanfu:1.1),outdoors
谁说全裸就不能表现中国风了?
[attach]3937613[/attach]
再删掉(Hanfu:1.1)试试,感觉变成韩国人了
[attach]3937614[/attach]
加回(Hanfu:1.1)。
这个的手指看起来还是有点别扭,看来得再加点关于手的负面提示词。
[attach]3937615[/attach]
加上了(sweat:1.5)测试出汗的效果,有点多了,像是被射到身上了
[attach]3937616[/attach]
改成(sweat:1.4)就好多了
[attach]3937617[/attach]
[attach]3937618[/attach]
改成(sweat:1.3)再看看,微汗就挺有感觉了
[attach]3937619[/attach]
用这个图的随机数种子再生成几张图,看看长相是不是相似
[attach]3937620[/attach]
[attach]3937621[/attach]
[attach]3937622[/attach]
[ 本帖最后由 内力煮火锅 于 2023-5-31 22:57(GMT+8) 编辑 ]
作者: scanroot 时间: 2023-3-24 12:46
听起来还是有点复杂,不过小姐姐做的是真漂亮
作者: cherry百分百 时间: 2023-3-24 13:13
执行新的安装指令的时候出现了错误:
Object ` /content/stable-diffusion-webui` not found.
fatal: not a git repository (or any of the parent directories): .git
python3: can't open file '/content/launch.py': [Errno 2] No such file or directory
[attach]3937629[/attach]
执行之前那个帖子的安装指令没有错误,正常启动
作者: 内力煮火锅 时间: 2023-3-24 13:16
引用:
原帖由 cherry百分百 于 2023-3-24 13:13 发表 [bbs=redirect.php?goto=findpost&pid=111736911&ptid=11593081]
[/bbs]
执行新的安装指令的时候出现了错误:
Object ` /content/stable-diffusion-webui` not found.
fatal: not a git repository (or any of the parent directories): .git
python3: can't open file '/content/launch.py ...
你是本地部署的还是用的帖子里的在线的平台?
作者: cherry百分百 时间: 2023-3-24 13:18
引用:
原帖由 内力煮火锅 于 2023-3-24 13:16 发表 [bbs=redirect.php?goto=findpost&pid=111736925&ptid=11593081]
[/bbs]
你是本地部署的还是用的帖子里的在线的平台?
在线平台啊,我再试一试吧,版主安装的时候没有报错吗?而且它报错的内容很奇怪,找不到安装包
作者: 内力煮火锅 时间: 2023-3-24 13:19
引用:
原帖由 cherry百分百 于 2023-3-24 13:18 发表 [bbs=redirect.php?goto=findpost&pid=111736934&ptid=11593081]
[/bbs]
在线平台啊
找到问题了,指令里有个标点符号打错了,现在已经修复了,你重新复制一遍,再试试。
作者: cherry百分百 时间: 2023-3-24 13:33
哈哈,我也猜是这种原因,刚才还在指令里面找,看和之前有哪些不同。
正常出图了,感谢版主![](images/smilies/default/lol.gif)
不知道是不是爆了显存,用1024*1024的时候一直出不了图
[ 本帖最后由 cherry百分百 于 2023-3-24 14:26(GMT+8) 编辑 ]
作者: 内力煮火锅 时间: 2023-3-24 13:39
引用:
原帖由 cherry百分百 于 2023-3-24 13:33 发表 [bbs=redirect.php?goto=findpost&pid=111737008&ptid=11593081]
[/bbs]
哈哈,我也猜是这种原因,刚才还在指令里面找,看和之前有哪些不同。
正常出图了,感谢版主![](images/smilies/default/lol.gif)
不知道是不是爆了显存,用1024*1024的时候一直出不了图 ...
是的,经过多次测试后发现,目前长和宽在800 X 800以内是安全的,可以适当增大其中一个数的同时减小另一个数。
另外出的图较大的话,会很快用完谷歌给你的免费额度,进入贤者时间。
作者: 1850482 时间: 2023-3-24 15:51
版主你好,我跑这个程序显示显示错误是
mv: cannot move '/content/chilloutmixni.safetensors' to '/content/stable-diffusion-webui/models/Stable-diffusion': No such file or directory
mv: cannot stat '/content/ulzzang-6500.pt': No such file or directory
/content/stable-diffusion-webui
python3: can't open file '/content/stable-diffusion-webui/launch.py': [Errno 2] No such file or directory
这个是什么原因呢
作者: 内力煮火锅 时间: 2023-3-24 16:15
引用:
原帖由 1850482 于 2023-3-24 15:51 发表 [bbs=redirect.php?goto=findpost&pid=111737498&ptid=11593081]
[/bbs]
版主你好,我跑这个程序显示显示错误是
mv: cannot move '/content/chilloutmixni.safetensors' to '/content/stable-diffusion-webui/models/Stable-diffusion': No such file or directory
mv: cannot stat '/conte ...
你再把安装指令完整复制一遍,然后再运行一遍试试看。
如果还是报错的话,就把第一行指令!pip install --upgrade fastapi==0.90.1删掉,再运行一遍试试看。
还不行的话,就把output输出项全复制过来我研究一下。
[ 本帖最后由 内力煮火锅 于 2023-3-24 16:18(GMT+8) 编辑 ]
作者: 1850482 时间: 2023-3-24 16:27
麻烦版主看一看是哪里的问题,删除第一行后可以生成网址,但是网址点不进去。而且No such file or directory这个报错还是存在
fatal: destination path 'stable-diffusion-webui' already exists and is not an empty directory.
fatal: destination path '/content/stable-diffusion-webui/extensions/stable-diffusion-webui-images-browser' already exists and is not an empty directory.
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1129 100 1129 0 0 1225 0 --:--:-- --:--:-- --:--:-- 1224
100 7346M 100 7346M 0 0 235M 0 0:00:31 0:00:31 --:--:-- 227M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1125 100 1125 0 0 1240 0 --:--:-- --:--:-- --:--:-- 1240
100 10155 100 10155 0 0 10330 0 --:--:-- --:--:-- --:--:-- 10330
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1159 100 1159 0 0 1275 0 --:--:-- --:--:-- --:--:-- 1273
100 144M 100 144M 0 0 103M 0 0:00:01 0:00:01 --:--:-- 307M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1165 100 1165 0 0 1284 0 --:--:-- --:--:-- --:--:-- 1283
100 144M 100 144M 0 0 105M 0 0:00:01 0:00:01 --:--:-- 105M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1169 100 1169 0 0 1281 0 --:--:-- --:--:-- --:--:-- 1280
100 144M 100 144M 0 0 86.8M 0 0:00:01 0:00:01 --:--:-- 195M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1169 100 1169 0 0 1297 0 --:--:-- --:--:-- --:--:-- 1296
100 144M 100 144M 0 0 104M 0 0:00:01 0:00:01 --:--:-- 275M
mkdir: cannot create directory ‘/content/stable-diffusion-webui/models/Lora’: File exists
mv: cannot stat '/content/koreanDollLikeness_v10.safetensors': No such file or directory
mv: cannot stat '/content/koreanDollLikeness_v15.safetensors': No such file or directory
mv: cannot stat '/content/japaneseDollLikeness_v10.safetensors': No such file or directory
mv: cannot stat '/content/taiwanDollLikeness_v10.safetensors': No such file or directory
mv: cannot stat '/content/chilloutmixni.safetensors': No such file or directory
mv: cannot stat '/content/ulzzang-6500.pt': No such file or directory
/content/stable-diffusion-webui
HEAD is now at 91c8d0dc Merge pull request #7231 from EllangoK/master
Python 3.9.16 (main, Dec 7 2022, 01:11:51)
[GCC 9.4.0]
Commit hash: 91c8d0dcfc9a5d46aec47fd3eb34c633c104b5bb
Installing xformers
Installing requirements for Web UI
Launching Web UI with arguments: --share --disable-safe-unpickle --no-half-vae --xformers --enable-insecure-extension-access
2023-03-24 08:22:19.150995: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-03-24 08:22:21.396615: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/lib64-nvidia
2023-03-24 08:22:21.396773: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/lib64-nvidia
2023-03-24 08:22:21.396812: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Loading weights [7234b76e42] from /content/stable-diffusion-webui/models/Stable-diffusion/chilloutmixni.safetensors
Creating model from config: /content/stable-diffusion-webui/configs/v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Applying xformers cross attention optimization.
Textual inversion embeddings loaded(1): ulzzang-6500
Model loaded in 29.2s (load weights from disk: 1.5s, create model: 1.3s, apply weights to model: 18.1s, apply half(): 1.2s, load VAE: 6.3s, move model to device: 0.6s).
Running on local URL: http://127.0.0.1:7860
Running on public URL: https://803c5747-603b-4bd3.gradio.live
This share link expires in 72 hours. For free permanent hosting and GPU upgrades (NEW!), check out Spaces: https://huggingface.co/spaces
作者: 内力煮火锅 时间: 2023-3-24 16:50
引用:
原帖由 1850482 于 2023-3-24 16:27 发表 [bbs=redirect.php?goto=findpost&pid=111737622&ptid=11593081]
[/bbs]
麻烦版主看一看是哪里的问题,删除第一行后可以生成网址,但是网址点不进去。而且No such file or directory这个报错还是存在
fatal: destination path 'stable-diffusion-webui' already exists and is not an empty d ...
哦,明白了,你是不是之前按照第一个教程运行过一遍,然后后来又按照这个教程的指令运行了一遍?
你可以把Colab的网页关掉,让它断开连接,然后再重新打开,然后你在右上角选择连接,然后按照教程里的方法,在设置里打开显卡。
然后再把全部的安装指令复制过去运行一遍。
作者: c7890 时间: 2023-3-24 19:00
太赞了
作者: 1850482 时间: 2023-3-24 21:27
引用:
原帖由 内力煮火锅 于 2023-3-24 16:50 发表 [bbs=redirect.php?goto=findpost&pid=111737685&ptid=11593081]
[/bbs]
哦,明白了,你是不是之前按照第一个教程运行过一遍,然后后来又按照这个教程的指令运行了一遍?
你可以把Colab的网页关掉,让它断开连接,然后再重新打开,然后你在右上角选择连接,然后按照教程里的方法,在设置里打开显卡。
然后再 ...
试了一下还是不行,还是一样的报错
Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/
Requirement already satisfied: fastapi==0.90.1 in /usr/local/lib/python3.9/dist-packages (0.90.1)
Requirement already satisfied: pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2 in /usr/local/lib/python3.9/dist-packages (from fastapi==0.90.1) (1.10.6)
Requirement already satisfied: starlette<0.24.0,>=0.22.0 in /usr/local/lib/python3.9/dist-packages (from fastapi==0.90.1) (0.23.1)
Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.9/dist-packages (from pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2->fastapi==0.90.1) (4.5.0)
Requirement already satisfied: anyio<5,>=3.4.0 in /usr/local/lib/python3.9/dist-packages (from starlette<0.24.0,>=0.22.0->fastapi==0.90.1) (3.6.2)
Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.9/dist-packages (from anyio<5,>=3.4.0->starlette<0.24.0,>=0.22.0->fastapi==0.90.1) (3.4)
Requirement already satisfied: sniffio>=1.1 in /usr/local/lib/python3.9/dist-packages (from anyio<5,>=3.4.0->starlette<0.24.0,>=0.22.0->fastapi==0.90.1) (1.3.0)
fatal: destination path 'stable-diffusion-webui' already exists and is not an empty directory.
fatal: destination path '/content/stable-diffusion-webui/extensions/stable-diffusion-webui-images-browser' already exists and is not an empty directory.
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1137 100 1137 0 0 4242 0 --:--:-- --:--:-- --:--:-- 4242
100 7346M 100 7346M 0 0 242M 0 0:00:30 0:00:30 --:--:-- 248M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1131 100 1131 0 0 4127 0 --:--:-- --:--:-- --:--:-- 4127
100 10155 100 10155 0 0 29692 0 --:--:-- --:--:-- --:--:-- 29692
mv: cannot stat '/content/chilloutmixni.safetensors': No such file or directory
mv: cannot stat '/content/ulzzang-6500.pt': No such file or directory
/content/stable-diffusion-webui
Python 3.9.16 (main, Dec 7 2022, 01:11:51)
[GCC 9.4.0]
Commit hash: 91c8d0dcfc9a5d46aec47fd3eb34c633c104b5bb
Installing xformers
Installing requirements for Web UI
Launching Web UI with arguments: --share --disable-safe-unpickle --no-half-vae --xformers --enable-insecure-extension-access
2023-03-24 13:22:44.857247: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-03-24 13:22:47.477889: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/lib64-nvidia
2023-03-24 13:22:47.478081: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/lib64-nvidia
2023-03-24 13:22:47.478104: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Loading weights [7234b76e42] from /content/stable-diffusion-webui/models/Stable-diffusion/chilloutmixni.safetensors
Creating model from config: /content/stable-diffusion-webui/configs/v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Applying xformers cross attention optimization.
Textual inversion embeddings loaded(1): ulzzang-6500
Model loaded in 31.4s (load weights from disk: 1.5s, create model: 1.3s, apply weights to model: 19.8s, apply half(): 1.1s, load VAE: 7.1s, move model to device: 0.6s).
Running on local URL: http://127.0.0.1:7860
这次网址都跑不出来了
作者: 内力煮火锅 时间: 2023-3-24 22:06
点这里打开文件管理
[attach]3938002[/attach]
把content文件夹里面除了sample_data以外的东西都删掉,只限于content文件夹里面的东西哈。
找不到content文件夹就点这个绿圈里的这个返回上一级文件夹按钮哈
[attach]3938003[/attach]
[ 本帖最后由 内力煮火锅 于 2023-3-24 22:08(GMT+8) 编辑 ]
作者: 1850482 时间: 2023-3-24 23:57
引用:
原帖由 内力煮火锅 于 2023-3-24 22:06 发表 [bbs=redirect.php?goto=findpost&pid=111738826&ptid=11593081]
[/bbs]
点这里打开文件管理
3938002
把content文件夹里面除了sample_data以外的东西都删掉,只限于content文件夹里面的东西哈。
找不到content文件夹就点这个绿圈里的这个返回上一级文件夹按钮哈
3938003 ...
多谢版主,现在程序可以跑了,但是又出了个问题,给了网址点进网站后是个空白页,不知道是啥问题。
Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/
Collecting fastapi==0.90.1
Downloading fastapi-0.90.1-py3-none-any.whl (56 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 56.2/56.2 KB 3.2 MB/s eta 0:00:00
Collecting starlette<0.24.0,>=0.22.0
Downloading starlette-0.23.1-py3-none-any.whl (64 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 64.5/64.5 KB 8.3 MB/s eta 0:00:00
Requirement already satisfied: pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2 in /usr/local/lib/python3.9/dist-packages (from fastapi==0.90.1) (1.10.6)
Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.9/dist-packages (from pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2->fastapi==0.90.1) (4.5.0)
Collecting anyio<5,>=3.4.0
Downloading anyio-3.6.2-py3-none-any.whl (80 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 80.6/80.6 KB 12.0 MB/s eta 0:00:00
Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.9/dist-packages (from anyio<5,>=3.4.0->starlette<0.24.0,>=0.22.0->fastapi==0.90.1) (3.4)
Collecting sniffio>=1.1
Downloading sniffio-1.3.0-py3-none-any.whl (10 kB)
Installing collected packages: sniffio, anyio, starlette, fastapi
Successfully installed anyio-3.6.2 fastapi-0.90.1 sniffio-1.3.0 starlette-0.23.1
Cloning into 'stable-diffusion-webui'...
remote: Enumerating objects: 17160, done.
remote: Total 17160 (delta 0), reused 0 (delta 0), pack-reused 17160
Receiving objects: 100% (17160/17160), 27.92 MiB | 38.79 MiB/s, done.
Resolving deltas: 100% (11999/11999), done.
Cloning into '/content/stable-diffusion-webui/extensions/stable-diffusion-webui-images-browser'...
remote: Enumerating objects: 143, done.
remote: Counting objects: 100% (34/34), done.
remote: Compressing objects: 100% (12/12), done.
remote: Total 143 (delta 25), reused 22 (delta 22), pack-reused 109
Receiving objects: 100% (143/143), 37.96 KiB | 1.41 MiB/s, done.
Resolving deltas: 100% (51/51), done.
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1131 100 1131 0 0 8440 0 --:--:-- --:--:-- --:--:-- 8440
100 7346M 100 7346M 0 0 230M 0 0:00:31 0:00:31 --:--:-- 222M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1125 100 1125 0 0 7120 0 --:--:-- --:--:-- --:--:-- 7120
100 10155 100 10155 0 0 43583 0 --:--:-- --:--:-- --:--:-- 43583
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1173 100 1173 0 0 9164 0 --:--:-- --:--:-- --:--:-- 9164
100 144M 100 144M 0 0 119M 0 0:00:01 0:00:01 --:--:-- 167M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1169 100 1169 0 0 9277 0 --:--:-- --:--:-- --:--:-- 9204
100 144M 100 144M 0 0 110M 0 0:00:01 0:00:01 --:--:-- 234M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1169 100 1169 0 0 9132 0 --:--:-- --:--:-- --:--:-- 9132
100 144M 100 144M 0 0 121M 0 0:00:01 0:00:01 --:--:-- 211M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1161 100 1161 0 0 9070 0 --:--:-- --:--:-- --:--:-- 9070
100 144M 100 144M 0 0 111M 0 0:00:01 0:00:01 --:--:-- 158M
/content/stable-diffusion-webui
Note: switching to '91c8d0d'.
You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.
If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:
git switch -c <new-branch-name>
Or undo this operation with:
git switch -
Turn off this advice by setting config variable advice.detachedHead to false
HEAD is now at 91c8d0dc Merge pull request #7231 from EllangoK/master
Python 3.9.16 (main, Dec 7 2022, 01:11:51)
[GCC 9.4.0]
Commit hash: 91c8d0dcfc9a5d46aec47fd3eb34c633c104b5bb
Installing gfpgan
Installing clip
Installing open_clip
Installing xformers
Cloning Stable Diffusion into repositories/stable-diffusion-stability-ai...
Cloning Taming Transformers into repositories/taming-transformers...
Cloning K-diffusion into repositories/k-diffusion...
Cloning CodeFormer into repositories/CodeFormer...
Cloning BLIP into repositories/BLIP...
Installing requirements for CodeFormer
Installing requirements for Web UI
Launching Web UI with arguments: --share --disable-safe-unpickle --no-half-vae --xformers --enable-insecure-extension-access
2023-03-24 15:47:28.932395: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-03-24 15:47:32.100381: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/lib64-nvidia
2023-03-24 15:47:32.100731: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/lib64-nvidia
2023-03-24 15:47:32.100760: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Calculating sha256 for /content/stable-diffusion-webui/models/Stable-diffusion/chilloutmixni.safetensors: 7234b76e423f010b409268386062a4111c0da6adebdf3a9b1a825937bdf17683
Loading weights [7234b76e42] from /content/stable-diffusion-webui/models/Stable-diffusion/chilloutmixni.safetensors
Creating model from config: /content/stable-diffusion-webui/configs/v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Downloading (…)olve/main/vocab.json: 100% 961k/961k [00:00<00:00, 7.02MB/s]
Downloading (…)olve/main/merges.txt: 100% 525k/525k [00:00<00:00, 4.48MB/s]
Downloading (…)cial_tokens_map.json: 100% 389/389 [00:00<00:00, 146kB/s]
Downloading (…)okenizer_config.json: 100% 905/905 [00:00<00:00, 372kB/s]
Downloading (…)lve/main/config.json: 100% 4.52k/4.52k [00:00<00:00, 640kB/s]
Applying xformers cross attention optimization.
Textual inversion embeddings loaded(1): ulzzang-6500
Model loaded in 78.4s (calculate hash: 45.7s, load weights from disk: 1.6s, create model: 2.7s, apply weights to model: 19.5s, apply half(): 1.2s, load VAE: 7.0s, move model to device: 0.6s).
Checkpoint chilloutmixni.safetensors [7234b76e42] not found; loading fallback chilloutmixni.safetensors [7234b76e42]
Running on local URL: http://127.0.0.1:7860
Running on public URL: https://b6cabd88-3076-4fb0.gradio.live
This share link expires in 72 hours. For free permanent hosting and GPU upgrades (NEW!), check out Spaces: https://huggingface.co/spaces
作者: 152853439712282 时间: 2023-3-25 00:48
请问有没方法将分辨率提升?目前768x1024都比较稳定,如果用高清修复拉高分辨率,画面很容易就崩了
作者: iGreed 时间: 2023-3-25 07:50
这个能做到两人吗?一人脱衣一人穿衣的效果,我试的要不两个都脱,要不两个都穿。而且同时两个人时,立绘的颜值也减半了
作者: 内力煮火锅 时间: 2023-3-25 08:09
引用:
原帖由 iGreed 于 2023-3-25 07:50 发表 [bbs=redirect.php?goto=findpost&pid=111740402&ptid=11593081]
[/bbs]
这个能做到两人吗?一人脱衣一人穿衣的效果,我试的要不两个都脱,要不两个都穿。而且同时两个人时,立绘的颜值也减半了
[bbs]viewthread.php?tid=11593536&page=1#pid111738567[/bbs]
我昨天看到这个帖子里有教两个人的提示词怎么写
作者: 内力煮火锅 时间: 2023-3-25 08:11
引用:
原帖由 152853439712282 于 2023-3-25 00:48 发表 [bbs=redirect.php?goto=findpost&pid=111739776&ptid=11593081]
[/bbs]
请问有没方法将分辨率提升?目前768x1024都比较稳定,如果用高清修复拉高分辨率,画面很容易就崩了
谷歌免费给的运算资源差不多就是这么大了。
要么就得自己架设,用自己的显卡了。
作者: 内力煮火锅 时间: 2023-3-25 08:21
可能得从下面几个方面逐一排查:
1. 确保运行指令后没有别的操作,出现网址以后也不要中断指令的运行,也不要刷新Colab的网页。
2. 不知道这个Colab是不是你新开的号,以前是不是运行过或安装过或设置过什么别的东西。如果不是新号,可以重新注册一个新号试试。
3. 也有可能本身安装的没问题但是生成的网页链接被当地网络运营商,翻墙软件,浏览器都拦截了。
下次发来输出项的时候先不要中断指令的运行或关闭网页,如果我正好在线,我可以试试我能不能打开,来判断你那边有没有被拦截。
另外,你说网址点进网站后是个空白页,是全白吗?空白页上有没有什么文字提示?
作者: haikui 时间: 2023-3-25 14:26
HEAD is now at 91c8d0dc Merge pull request #7231 from EllangoK/master
Python 3.9.16 (main, Dec 7 2022, 01:11:51)
[GCC 9.4.0]
Commit hash: 91c8d0dcfc9a5d46aec47fd3eb34c633c104b5bb
Traceback (most recent call last):
File "/content/stable-diffusion-webui/launch.py", line 359, in <module>
prepare_environment()
File "/content/stable-diffusion-webui/launch.py", line 271, in prepare_environment
run_python("import torch; assert torch.cuda.is_available(), 'Torch is not able to use GPU; add --skip-torch-cuda-test to COMMANDLINE_ARGS variable to disable this check'")
File "/content/stable-diffusion-webui/launch.py", line 129, in run_python
return run(f'"{python}" -c "{code}"', desc, errdesc)
File "/content/stable-diffusion-webui/launch.py", line 105, in run
raise RuntimeError(message)
RuntimeError: Error running command.
Command: "/usr/bin/python3" -c "import torch; assert torch.cuda.is_available(), 'Torch is not able to use GPU; add --skip-torch-cuda-test to COMMANDLINE_ARGS variable to disable this check'"
Error code: 1
stdout: <empty>
stderr: Traceback (most recent call last):
File "<string>", line 1, in <module>
AssertionError: Torch is not able to use GPU; add --skip-torch-cuda-test to COMMANDLINE_ARGS variable to disable this check
楼主能不能帮看下哪里出了问题
作者: 内力煮火锅 时间: 2023-3-25 16:19
引用:
原帖由 haikui 于 2023-3-25 14:26 发表 [bbs=redirect.php?goto=findpost&pid=111741846&ptid=11593081]
[/bbs]
HEAD is now at 91c8d0dc Merge pull request #7231 from EllangoK/master
Python 3.9.16 (main, Dec 7 2022, 01:11:51)
[GCC 9.4.0]
Commit hash: 91c8d0dcfc9a5d46aec47fd3eb34c633c104b5bb
Traceback (most rec ...
最后一行改成复制内容到剪贴板
代码:
!COMMANDLINE_ARGS="--share --disable-safe-unpickle --no-half-vae --xformers --reinstall-xformers --enable-insecure-extension-access --skip-torch-cuda-test" REQS_FILE="requirements.txt" python launch.py
然后再运行试试
作者: 152853439712282 时间: 2023-3-25 20:51
引用:
原帖由 内力煮火锅 于 2023-3-25 08:11 发表 [bbs=redirect.php?goto=findpost&pid=111740468&ptid=11593081]
[/bbs]
谷歌免费给的运算资源差不多就是这么大了。
要么就得自己架设,用自己的显卡了。
我现在就是在本地部署的,用自己显卡,3070
作者: 内力煮火锅 时间: 2023-3-25 21:10
引用:
原帖由 152853439712282 于 2023-3-25 20:51 发表 [bbs=redirect.php?goto=findpost&pid=111743343&ptid=11593081]
[/bbs]
我现在就是在本地部署的,用自己显卡,3070
那我估计就是得多打磨打磨提示词和负面提示词做实验看看怎么样能正常出更大的图了。
作者: wyanyan 时间: 2023-3-25 22:46
第一次看到用AI修整图片的程序代码,虽然看不太懂,但真是大神啊。
作者: 我心依然3 时间: 2023-3-25 22:50
给人感觉好复杂的样子,很难学会呀,可以试试,软件是用哪一种好?
作者: wert25 时间: 2023-3-26 04:42
我也想试试,就是感觉有点复杂,要研究研究
作者: cflwd 时间: 2023-3-26 15:59
受教受教,现在的AI技术实在太强大了,学好了这个,好像能满足所有的幻想了。
作者: anubisw198710 时间: 2023-3-26 21:47
楼主,求一个科学上网靠谱的软件,有手机用的吗?
作者: swordolive 时间: 2023-3-27 21:14
牛逼的起飞的帖子,真是震惊到我了。
作者: yong778 时间: 2023-3-27 22:41
回去试试
作者: susuhao 时间: 2023-3-27 23:10
Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/
Requirement already satisfied: fastapi==0.90.1 in /usr/local/lib/python3.9/dist-packages (0.90.1)
Requirement already satisfied: pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2 in /usr/local/lib/python3.9/dist-packages (from fastapi==0.90.1) (1.10.7)
Requirement already satisfied: starlette<0.24.0,>=0.22.0 in /usr/local/lib/python3.9/dist-packages (from fastapi==0.90.1) (0.23.1)
Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.9/dist-packages (from pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2->fastapi==0.90.1) (4.5.0)
Requirement already satisfied: anyio<5,>=3.4.0 in /usr/local/lib/python3.9/dist-packages (from starlette<0.24.0,>=0.22.0->fastapi==0.90.1) (3.6.2)
Requirement already satisfied: sniffio>=1.1 in /usr/local/lib/python3.9/dist-packages (from anyio<5,>=3.4.0->starlette<0.24.0,>=0.22.0->fastapi==0.90.1) (1.3.0)
Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.9/dist-packages (from anyio<5,>=3.4.0->starlette<0.24.0,>=0.22.0->fastapi==0.90.1) (3.4)
fatal: destination path 'stable-diffusion-webui' already exists and is not an empty directory.
fatal: destination path '/content/stable-diffusion-webui/extensions/stable-diffusion-webui-images-browser' already exists and is not an empty directory.
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1129 100 1129 0 0 4003 0 --:--:-- --:--:-- --:--:-- 4003
100 7346M 100 7346M 0 0 196M 0 0:00:37 0:00:37 --:--:-- 210M
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1127 100 1127 0 0 3954 0 --:--:-- --:--:-- --:--:-- 3940
100 10155 100 10155 0 0 27821 0 --:--:-- --:--:-- --:--:-- 27821
mv: cannot stat '/content/chilloutmixni.safetensors': No such file or directory
mv: cannot stat '/content/ulzzang-6500.pt': No such file or directory
/content/stable-diffusion-webui
Python 3.9.16 (main, Dec 7 2022, 01:11:51)
[GCC 9.4.0]
Commit hash: 91c8d0dcfc9a5d46aec47fd3eb34c633c104b5bb
Installing xformers
Installing requirements for Web UI
Launching Web UI with arguments: --share --disable-safe-unpickle --no-half-vae --xformers --enable-insecure-extension-access
Warning: caught exception 'No CUDA GPUs are available', memory monitor disabled
2023-03-27 15:04:56.603685: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-03-27 15:04:59.487051: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
2023-03-27 15:04:59.487382: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/lib/python3.9/dist-packages/cv2/../../lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
2023-03-27 15:04:59.487414: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Loading weights [7234b76e42] from /content/stable-diffusion-webui/models/Stable-diffusion/chilloutmixni.safetensors
Creating model from config: /content/stable-diffusion-webui/configs/v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
loading stable diffusion model: RuntimeError
Traceback (most recent call last):
File "/content/stable-diffusion-webui/webui.py", line 103, in initialize
modules.sd_models.load_model()
File "/content/stable-diffusion-webui/modules/sd_models.py", line 406, in load_model
sd_hijack.model_hijack.hijack(sd_model)
File "/content/stable-diffusion-webui/modules/sd_hijack.py", line 107, in hijack
self.optimization_method = apply_optimizations()
File "/content/stable-diffusion-webui/modules/sd_hijack.py", line 39, in apply_optimizations
if cmd_opts.force_enable_xformers or (cmd_opts.xformers and shared.xformers_available and torch.version.cuda and (6, 0) <= torch.cuda.get_device_capability(shared.device) <= (9, 0)):
File "/usr/local/lib/python3.9/dist-packages/torch/cuda/__init__.py", line 357, in get_device_capability
prop = get_device_properties(device)
File "/usr/local/lib/python3.9/dist-packages/torch/cuda/__init__.py", line 371, in get_device_properties
_lazy_init() # will define _get_device_properties
File "/usr/local/lib/python3.9/dist-packages/torch/cuda/__init__.py", line 229, in _lazy_init
torch._C._cuda_init()
RuntimeError: No CUDA GPUs are available
Stable diffusion model failed to load, exiting
楼主能帮我看看哪里的问题,谢谢您,我是苹果电脑IMAC
[ 本帖最后由 susuhao 于 2023-3-27 23:11(GMT+8) 编辑 ]
作者: 内力煮火锅 时间: 2023-3-28 10:12
你看看是不是下面这个原因:引用:
原帖由 小小丁丁 于 2023-3-25 16:53 发表 [bbs=redirect.php?goto=findpost&pid=111742384&ptid=11590974]
[/bbs]
这里做个提示,MacBook的M芯片系列电脑不用尝试了,它是ARM架构的GPU,不支持CUDA GPU,安装不了
另外,如果你不是本地部署,而是在线部署的话,那就跟自己的电脑的型号没关系了,这种情况就要看一下有没有在colab的设置里把白嫖谷歌的显卡算力资源打开。
[ 本帖最后由 内力煮火锅 于 2023-3-28 10:15(GMT+8) 编辑 ]
作者: asgardv 时间: 2023-3-28 11:27
楼主,我想问一下,提示词是随便什么英文么?还是有固定的提示词。另外加双括号是什么意思,是优先级提高还是变低啊,我弄了几个词都没什么太大的效果,而且我生成的图像都是半身的,我想要全身的。
作者: 内力煮火锅 时间: 2023-3-28 15:45
引用:
原帖由 asgardv 于 2023-3-28 11:27 发表 [bbs=redirect.php?goto=findpost&pid=111755215&ptid=11593081]
[/bbs]
楼主,我想问一下,提示词是随便什么英文么?还是有固定的提示词。另外加双括号是什么意思,是优先级提高还是变低啊,我弄了几个词都没什么太大的效果,而且我生成的图像都是半身的,我想要全身的。 ...
理论上所有的词都是可以的,看AI对语言文字的理解程度了。
圆括号是加权,方括号是降权,我的前一篇教程的最后简单介绍了一下提示词的语法规则。
提示词没什么用有可能是权重不够,也可能是AI对这个词不太理解,还可能是提示词太多了AI照顾不过来了。
想要下半身你的提示词里就要提到下半身,你可以描述一个身体部位,或者描述一个姿势。
全身也是一样,你要描述一下身体的样子,或者一个涉及到全身的姿势。
作者: frankwin05 时间: 2023-3-28 17:27
要是都像楼主这样出教程,我TM早就博士毕业了!
跪赞!
作者: lztitan 时间: 2023-3-28 18:49
楼主能否开的帖子教教我们怎么弄真人头像啊,我对这个比较好奇。
作者: ac-fish 时间: 2023-3-30 22:42
打开colab页面后,在左上方有个修改,点那里,配置下笔记本,选择GPU,在运行代码安装应该就可以了
作者: 滴水 时间: 2023-4-2 08:47
大佬真多,这肯定不是业余的,主业绝对NB,萌新瑟瑟发抖,搞个VAM我都哆哆嗦嗦,编程还是算了吧
作者: zz4816 时间: 2023-4-2 15:00
大佬,你这样干的话,那些福利鸡就要重操旧业回到KTV,夜总会了,大佬你这可积了大德了!
作者: Ahaa67 时间: 2023-4-3 09:46
厉害了,不过看了半天还是学不会,我这技术只能膜拜大神了
作者: hellomickey 时间: 2023-4-3 13:19
想请问下这个方法可以用自定义的人像照片吗?比如换脸一类?
作者: llouies 时间: 2023-4-3 20:52
还没试过,但是不论是楼主还是楼内大佬们的作品最后一张微笑表情下都会陷入恐怖谷。。可能露齿微笑缺乏面部肌肉细节?
作者: suningdianqi 时间: 2023-4-3 22:46
人体不协调,特别过分追求超大的乳房,虽然i满足了一些人的嗜好,但有损东方美人的整体美,有点遗憾。
作者: dxyyygsliu 时间: 2023-4-5 17:56 标题: 版主,输入安装指令后结果如下,安装失败,请指导一下
RuntimeError: Detected that PyTorch and torchvision were compiled with different CUDA versions. PyTorch has CUDA Version=11.7 and torchvision has CUDA Version=11.8. Please reinstall the torchvision that matches your PyTorch install.
[ 本帖最后由 dxyyygsliu 于 2023-4-5 18:06(GMT+8) 编辑 ]
作者: 内力煮火锅 时间: 2023-4-5 19:04
引用:
原帖由 dxyyygsliu 于 2023-4-5 17:56 发表 [bbs=redirect.php?goto=findpost&pid=111794229&ptid=11593081]
[/bbs]
RuntimeError: Detected that PyTorch and torchvision were compiled with different CUDA versions. PyTorch has CUDA Version=11.7 and torchvision has CUDA Version=11.8. Please reinstall the torchvision th ...
已更新安装指令,你再试一下复制全部的安装指令,再运行一遍试试,重新运行前记得先重新连接一下Colab(重启浏览器或在右上角小箭头下拉菜单处选择删除运行时/delete runtime)。
作者: zengyuhuang 时间: 2023-4-5 23:48
也不知道是不是电脑配置问题,总是跑不动图
作者: 满园春色2 时间: 2023-4-7 20:04 标题: 厉害
大神的技术真是神乎其神,一饱眼福
作者: kidme1314 时间: 2023-4-8 02:21
这让我想起了B站做手机屏保的那位大神,分享有那味了,而且开头提醒的那一句话真的很暖心
作者: feiyan1025 时间: 2023-4-8 14:37
http://23.225.255.95/forum/attachments/month_2303/20230324_c8e4f9a70d2964efd465A864wpjfkT01.png
大佬 这张图很喜欢 我根据大佬的步骤生成不出这个姿势 可否告知详细数据
作者: 内力煮火锅 时间: 2023-4-9 10:10
引用:
原帖由 feiyan1025 于 2023-4-8 14:37 发表 [bbs=redirect.php?goto=findpost&pid=111809331&ptid=11593081]
[/bbs]
![](attachments/month_2303/20230324_c8e4f9a70d2964efd465A864wpjfkT01.png)
大佬 这张图很喜欢 我根据大佬的步骤生成不出这个姿势 可否告知详细数据
...
把图丢进webUI的image info里面,就能看到所有参数。
或者丢到这个网站里也可以:
https://spell.novelai.dev/
作者: 3003323 时间: 2023-4-16 21:38
都是高手,我的笔记 本提示不支持,说是显示不是navdia的,教程不是说不要用的本地显卡吗?服了
作者: 内力煮火锅 时间: 2023-4-16 22:15
引用:
原帖由 3003323 于 2023-4-16 21:38 发表 [bbs=redirect.php?goto=findpost&pid=111851476&ptid=11593081]
[/bbs]
都是高手,我的笔记 本提示不支持,说是显示不是navdia的,教程不是说不要用的本地显卡吗?服了
你按照教程里的方法,用Colab运行这个,对本地显卡就没有要求了。
作者: tgyd277 时间: 2023-4-18 12:03
认真学习了一遍可惜不是干这个的材料,搞不懂!只有拜托各路大师费心费神AI出来。光板板屄搞得太多了,添点毛毛更显性感。
作者: 吸血魔兽 时间: 2023-4-22 06:48
感谢大佬,教程真的很是详细能学不少东西。对新人帮助很大
作者: iamhixitele 时间: 2023-4-22 07:04
引用:
原帖由 内力煮火锅 于 2023-4-16 22:15 发表 [bbs=redirect.php?goto=findpost&pid=111851617&ptid=11593081]
[/bbs]
你按照教程里的方法,用Colab运行这个,对本地显卡就没有要求了。
远程部署还是上auto dl租显卡吧,一小时1.5元就能玩的很花了,b站有教程
作者: a8909257 时间: 2023-11-21 01:32
大佬 你的代码还能用吗 我试了一下好像不太行呢
作者: diligite 时间: 2024-2-3 20:05
作者: binghudeyu 时间: 2024-2-7 23:32
对这个很感兴趣,但是看各位大神说得好复杂,只能望洋兴叹了。
好在现在科技日新月异,期待尽快出傻瓜版的,大神们能否把它弄成一个常规软件安装即可呢?制作图片的速度慢点儿也不要紧,哪怕一个小时一张都可以的,让它在后台慢慢跑就是了、
欢迎光临 SiS001! Board - [第一会所 关闭注册] (https://the001.net/bbs1/) |
Powered by Discuz! 7.2 |