search

WSH032 / kohya-config-webui

A WebUI for making config files used by kohya_sd_script. 一个用于生成kohya-ss训练脚本使用的toml配置文件的WebUI
Apache License 2.0
94 stars 4 forks source link

python setup.py egg_info did not run successfully 和 Encountered error while generating package metadata. #6

Closed 2949984428 closed 1 year ago

2949984428 commented 1 year ago

Cloning into '/content/sd-scripts'... remote: Enumerating objects: 2808, done. remote: Counting objects: 100% (1021/1021), done. remote: Compressing objects: 100% (276/276), done. remote: Total 2808 (delta 842), reused 847 (delta 745), pack-reused 1787 Receiving objects: 100% (2808/2808), 3.28 MiB | 4.97 MiB/s, done. Resolving deltas: 100% (1922/1922), done. Cloning into '/content/kohya-config-webui'... remote: Enumerating objects: 118, done. remote: Counting objects: 100% (118/118), done. remote: Compressing objects: 100% (91/91), done. remote: Total 118 (delta 58), reused 53 (delta 22), pack-reused 0 Receiving objects: 100% (118/118), 67.27 KiB | 13.45 MiB/s, done. Resolving deltas: 100% (58/58), done. torch安装中 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 108.2/108.2 MB 8.2 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 123.8/123.8 MB 8.0 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 123.6/123.6 MB 7.9 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 50.9/50.9 MB 12.9 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 292.5/292.5 kB 33.1 MB/s eta 0:00:00 error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully. │ exit code: 1 ╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip. Preparing metadata (setup.py) ... error error: metadata-generation-failed

× Encountered error while generating package metadata. ╰─> See above for output.

note: This is an issue with the package mentioned above, not pip. hint: See above for details. torch安装完成 kohya依赖安装中

WSH032 commented 1 year ago

收到。 colab昨天更新了torch2.0.0 -> torch2.0.1， 但是xformer和它不兼容现在，所以导致了上述问题的发生。

目前有两个解决方案：

降级torch

-!pip -q install torch torchvision xformers triton
+!pip -q install torch==2.0.0 torchvision xformers triton

缺点是安装时间多了3~4分钟

训练时不使用xfomer

torch2.0.1及相关依赖colab已经安装好了，什么代码也不用动，训练时候别开xformer就行 缺点是显存占用变高

我一会更新的时候会采用第一种方法。 你可能会注意到安装时会有一些红色警告，例如 torchaudio 2.0.2+cu118 requires torch==2.0.1, but you have torch 2.0.0 which is incompatible. 这个不影响训练（至少我没观察到）

2949984428 commented 1 year ago

感谢 另外 b站求继续更新