Type | |
Stats | 4,424 |
Reviews | |
Published | Apr 24, 2023 |
Base Model | |
Trigger Words | wearing mermaid dress wedding dress back view movie premier gala standing on the red carpet |
Hash | AutoV2 1AD580E189 |
Get early access to build and test build, be able to try all epochs and test them by yourself on Patreon or contact me for support on Discord
To support me you can buy me a coffe at https://ko-fi.com/antonioriolo
Training settings:
40 images, 15 steps, 12 epochs, you find many epochs (with test images using plot xy script) on my Patreon while supporting me.
Recommended settings:
Make sure you are aware on the usage instructions of LORA
Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)
CFG Scale : 5 to 9
LORA weight for txt2img: reccomended 0.2-0.6 (0.3 best results in my experiments)
Keys: wearing mermaid dress, movie premier gala, standing on the red carpet, wedding dress, back view
Adviced Image sIze: 512x768, 512x912
this model isn't hard to use even if it doesn't always give what you expect.
My advice is to use always higher height in the image to show all the dress and use keywords like full body.
You might want a mermaid wedding dress in that case you can just write "wearing mermaid dress, wedding dress"
Also, you might want to pick the dress color, in that case just specify it, the training images have a lot of variety, SD should be able to understand any dress color.
Many of the images i posted have DPM Karras but any DPM++ gives better result usually.
Enjoy!
If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: https://www.patreon.com/betterCuteAsians