-
### Is there an existing issue for this?
- [X] I have searched the existing issues and checked the recent builds/commits
### What happened?
I trained a lora on SDXL with this diffusers script: http…
-
Recently no longer able to train. Vram usage for simple lora bs1 with gradientcheckpointing has gone above 4090 capacity. Have optimisers on but I get either OOO message on Kohya v23 or 50 hours train…
-
When using GUI to change Lora Type even when selecting other types like locon and adding Conv values, tomol always comes out as standard lora networks.lora
-
I'm using a RTX 2080 Ti graphic card with 22G VRAM (second hand, self-modified). My previous LoRAs are trained in Sep'23 using a RTX3090 with 24G VRAM and it worked fine with fairly satisfying outcome…
-
error is:
INFO [Dataset 0] config_util.py:571
…
xddun updated
5 months ago
-
![image](https://github.com/bmaltais/kohya_ss/assets/151836533/7cc40f86-25e6-4fda-821d-043391251f46)
-
I've tried switching it from adam b bit to adafactor, still didn't work. Halllp!
-----------------------------
10:19:08-855002 INFO Version: v22.6.2
10:19:08-862468 INFO nVidia toolki…
-
When trying to install requirements with the CLI command`pip install -r requirements`, I got
```
ERROR: Invalid requirement: '_libgcc_mutex=0.1=main' (from line 4 of requirements.txt)
Hint: = is…
-
As the title suggests
Can kohya_ss's WebUI not be merged into Stable Diffusion and can it only exist as a standalone WebUI page?
-
### What happened and why was it ungood?
When I use (--unet-in-fp8-e4m3fn) in the command line, the following error occurs, and I cannot generate images.
This error only occurs when using the contr…