After a gap of two years, I trained using 20,000 images, and the SD lightning model.
From this I can provide this nice LoRA up weights. Feel free to use.
This repo is for testing.
I need your responses and opinons.
RealVisXL V4.0 V4.0 Lightning (BakedVAE) is used
Thus, Hires fix is a must! 2.0 scale, 0.6 denoise.
I may change the base model, current one have difficulty in generate a full body picture. (I suspect they only train the face...)
v2.1 using the larger base version, pls notice.
Plan:
Testing Optimizing Parameters (Currently train unet only)
Change base model
How to remove water mark at preprocess? (May need help)
DPO + LoRA (Large amount of code, it takes time)
GPU resources is still needed.
Description
FAQ
Comments (5)
trained using a lightning model? That always failed for me.
and why? doesn't training against the base model and then using a lightning model have the same results?
Lightning means less step. Checkpoint size is same as xl base one. I train from lightning checkpoint using sdxl script.
https://arxiv.org/html/2402.13929v1
SD lightning is a Diffusion Distillation. A different checkpoint.
@wumoe I know, but whenever I tried to train against a lightning checkpoint, I didn't get any training results. I always trained against the checkpoint without lightning, and then used the extracted LoRA with the lightning checkpoint (or the lightning LoRA)
Why does this work for you?
@jAC3KMdz3M5H may check the parameters, i didn't do much special.