https://github.com/deepghs/waifuc/tree/main
The model data source is derived from Waifuc, a highly robust tool for crawling, filtering, and processing training sets. With the aid of the Waifuc tool and a single 5600g machine, it is possible to efficiently crawl and process over 3000 training images within a span of 2 hours. Training the character "Lora" on this extensive dataset yields significant performance advantages compared to situations with limited data availability.
↑ Translation from gpt.
My english is not good and can‘t directly show my thoughts.
waifuc is really a powerful and fantasy tool.
Please use the latest v3 ver, the v1 v2 ver were my early attempts with train.
the v3 lora is train base 2000+ images &4000+ stages .
other older lora are being gradually reset
tag in page.
other in metainfo
recommand use anime style model
Description
FAQ
Comments (3)
hi, bro, i'm the author of waifuc project, leader of deepghs team. u tipped me for waifuc before. and i would ask are u satisfied with waifuc? if u need help with waifuc or waifuc simply disappointted u, just tell us, we are open to all kinds of feedback to improve this project. thank u, have a nice day.
I am quite satisfied with waifuc, especially since CCIP has solved almost all of the issues. I can easily use waifuc by writing a simple script in Python. Thank you and your team for creating such an outstanding work.
@kitarz glad that u like it. if ur facing some issues, just notify me, or open an issue on github repo: https://github.com/deepghs/waifuc/issues
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.