1.0.4

Kohya SS

Version 1.0.4

Included in this Template

• Ubuntu 22.04 LTS
• CUDA 11.8
• Python 3.10.6
• Kohya_ss v21.7.16
• Torch 2.0.1
• xformers 0.0.20

Ports

Port	Description
3000	Kohya_ss
8888	Jupyter Lab

Environment Variables

Variable	Description	Default
JUPYTER_PASSWORD	Password for Jupyter Lab	Jup1t3R!
DISABLE_AUTOLAUNCH	Disable Web UIs from launching automatically	enabled

Logs

Kohya SS creates a log files, and you can tail the log file
instead of killing the services to view the logs

Application	Log file
Kohya SS	/workspace/logs/kohya_ss.log

For example:

tail -f /workspace/logs/kohya_ss.log

Jupyter Lab

If you wish to use the Jupyter lab, you must set
the JUPYTER_PASSWORD environment variable in the
Template Overrides configuration when deploying
your pod.

General

Note that this does not work out of the box with
encrypted volumes!

This is a custom packaged template for the Kohya_ss
Web UI.

I do not maintain the code for this repo,
I just package everything together so that it is
easier for you to use.

If you need help with settings, etc. You can feel free
to ask me, but just keep in mind that I am not an expert
at Kohya_ss! I'll try my best to help, but the
RunPod community or Automatic/Kohya_ss communities
may be better at helping you.

Please wait until the GPU Utilization % is 0 before
attempting to connect. You will likely get a 502 error
before that as the pod is still getting ready to be used.

Using your own models

The best ways to get your models onto your pod is
by using runpodctl or by uploading them to Google
Drive or other cloud storage and downloading them
to your pod from there.

Uploading to Google Drive

If you're done with the pod and would like to send
things to Google Drive, you can use this colab to do it
using runpodctl. You run the runpodctl either in
a web terminal (found in the pod connect menu), or
in a terminal on the desktop.