doodad-v1-1

作者 doohickey · 下载 0 · 点赞 3 · 更新 2023-11-01

stable-diffusion text-to-image

Doodad

This is the 1st part of a 2 (3?) part project meant to be used with Doohickey

The style was trained with Dreambooth-Stable and is used with "imv" (an uncommonly used token recommended by the author I think) and it mixes especially well with the style included in Doohickey. It's a finetuned version of the Trinart-Waifu-diffusion-50-50 included in this organization's models.

Parameter Value
resolution 512
train_batch_size 1
gradient_accumulation_steps 2
learning_rate 5e-6
num_class_images 120
max_train_steps 1200

Example outputs:

"portal into another dimension" doodad-v1-1 预览 1

"portrait of a dying god" doodad-v1-1 预览 2

"photograph" doodad-v1-1 预览 3

"The alchemist's laboratory by Greg Rutkowski and Claude Monet, oil on canvas" doodad-v1-1 预览 4