Software Engineering Courses (SECourses)

SEC

Software Engineering Courses (SECourses)

Discuss, ask and learn Stable Diffusion, DreamBooth, Programming, Software Engineering, C#, ASP.NET, MSSQL and more

Join

script-updates-and-info

toros-uni

flux-stable-diffusion-such

programming

tips-info

ai

technology

server-admin

cyber-security

dreambooth-training

work-on-next-list

announcements

actually on the contrary you can have

actually on the contrary you can have single concept. from concept we mean single subject

which dataset where? which captions?

which dataset where? which captions?

I need some help. I trained a flux LoRA

I need some help. I trained a flux LoRA with highest training. So the text encoder was tuned as well. When i place the text encoder in the SwarmUI/Models/clip path. It re-downloads with the name t5xxl_enconly. ...

Anyone have experience creating a

Anyone have experience creating a finetuned model that is trained on a logo and can generate apparel / clothing with that logo?

Interesting! I will experiment with

Interesting! I will experiment with AdamW today.

hello everyone I need your help!!! I am

hello everyone I need your help!!! I am running the latest kohya_ss for SDXL Lora training with 26 training images * 8 repeats which should be around 208 training images with repeats but while the script recognizes that (INFO 208 train images with repeating. ) it loads double images and as a result I get double total training steps... Anyone else got similar issue?

are you testing the lora on the same

are you testing the lora on the same base model?