Type | |
Stats | 8,520 |
Reviews | |
Published | Jan 24, 2023 |
Base Model | |
Training | Steps: 2,000 |
Trigger Words | COLR_001 |
Hash | AutoV2 EE10B26E6C |
About model
A simple model trained on a custom dataset containing over 100 coloring book type images.
If you enjoy this model and would like me to improve on it, buy me a coffe ā
Installation:
Download both the ckpt and yaml files. Ensure that the same naming pattern is used and copy them under models/Stable-Diffusion path in your local/cloud SD installation. Stable Diffusion 2.1 is required for the model to work correctly.
Black images issue:
2.1 models need to have a web-ui config modified - if you are getting black images - go to your config file and add to COMMANDLINE_ARGS= --no-half - potentially it could work with --xformers instead (if supported). This line might slow your generations a bit but will not affect negatively your output.
Prompt suggestion:
bichon havanese wearing sunglasses COLR_001, (((white background))), coloring book, line art, high resolution, black and white, colorless
Negative: ((watermark)), (text), color, shading, gradient, shadows, transparency, noisy, blurred