Stable Diffusion 使用安装

605 阅读2分钟

Stable Diffusion是一个文本到图像的潜在扩散模型,由CompVis、Stability AI和LAION的研究人员和工程师创建。它使用来自LAION-5B数据库子集的512x512图像进行训练。使用这个模型,可以生成包括人脸在内的任何图像,因为有开源的预训练模型,所以我们也可以在自己的机器上运行它。

一、本地安装

Tip: Python 版本为 3.10.6

 git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
 ​
 cd stable-diffusion-webui/
 ​
 # 请切换Python venv, 魔法上网, 下载时间比较长
 git config --global http.proxy http://127.0.0.1:10887
 pip install --upgrade pip
 export http_proxy=http://127.0.0.1:10887;export https_proxy=http://127.0.0.1:10887;
 sh ./webui.sh

二、安装报错汇总

2.1 安装gfpgan报错

 Installing gfpgan
 Traceback (most recent call last):
   File "/Users/iarno/Devspace/chatgpt/stable-diffusion-webui/launch.py", line 355, in <module>
     prepare_environment()
   File "/Users/iarno/Devspace/chatgpt/stable-diffusion-webui/launch.py", line 263, in prepare_environment
     run_pip(f"install {gfpgan_package}", "gfpgan")
   File "/Users/iarno/Devspace/chatgpt/stable-diffusion-webui/launch.py", line 129, 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 "/Users/iarno/Devspace/chatgpt/stable-diffusion-webui/launch.py", line 97, in run
     raise RuntimeError(message)
 RuntimeError: Couldn't install gfpgan.
 Command: "/Users/iarno/Devspace/chatgpt/stable-diffusion-webui/venv/bin/python3.10" -m pip install git+https://github.com/TencentARC/GFPGAN.git@8d2447a2d918f8eba5a4a01463fd48e45126a379 --prefer-binary
 Error code: 2
 stdout: Collecting git+https://github.com/TencentARC/GFPGAN.git@8d2447a2d918f8eba5a4a01463fd48e45126a379
   Cloning https://github.com/TencentARC/GFPGAN.git (to revision 8d2447a2d918f8eba5a4a01463fd48e45126a379) to /private/var/folders/qw/4fkzpn79447fkgjgk19pq_6h0000gn/T/pip-req-build-0i1rio01
   
   
 stderr:   Running command git clone --filter=blob:none --quiet https://github.com/TencentARC/GFPGAN.git /private/var/folders/qw/4fkzpn79447fkgjgk19pq_6h0000gn/T/pip-req-build-0i1rio01
   Running command git rev-parse -q --verify 'sha^8d2447a2d918f8eba5a4a01463fd48e45126a379'
   Running command git fetch -q https://github.com/TencentARC/GFPGAN.git 8d2447a2d918f8eba5a4a01463fd48e45126a379
   Running command git checkout -q 8d2447a2d918f8eba5a4a01463fd48e45126a379

解决方案:

 git config --global http.proxy http://127.0.0.1:10887

2.2 未下载模型导致服务无法运行

 No checkpoints found. When searching for checkpoints, looked at:
  - file /Users/liuli/Devspace/chatgpt/stable-diffusion-webui/model.ckpt
  - directory /Users/liuli/Devspace/chatgpt/stable-diffusion-webui/models/Stable-diffusion
 Can't run without a checkpoint. Find and place a .ckpt or .safetensors file into any of those locations. The program will exit.

解决方案:

下载模型, 并将下载的模型放到 stable-diffusion-webui/models/Stable-diffusion 目录中。

三、使用Colab搭建

Colab 免费时长为12小时

本地模型生成图片会受限于💻配置问题, 推荐使用谷歌免费 Colab 服务搭建, 避免多次下载 Model 模型, 可以考虑使用挂载谷歌云盘的方式(提前将下载好的Model上传至谷歌云盘)。

3.1 安装教程:

xhslink.com/6lTt8o

3.2 修改项

挂载云盘:

image-20230416200116930

拷贝云盘模型:

image-20230416200311072

四、参考

github.com/AUTOMATIC11…

github.com/camenduru/s…

xhslink.com/6lTt8o

www.bilibili.com/video/BV1vX…

www.bilibili.com/video/BV1fL…