Type | |
Stats | 76 388 |
Reviews | (26) |
Published | Jul 30, 2024 |
Base Model | |
Training | Steps: 418 Epochs: 10 |
Usage Tips | Strength: 0.8 |
Trigger Words | kyle_hyde formal black suit, white collared shirt, black necktie blue jacket, white collared shirt, black necktie brown hair, stubble, gray eyes, 34 years old |
Hash | AutoV2 F077A3BF5E |
Kyle Hyde from the visual novel Hotel Dusk.
Activation tag: kyle_hyde
Game 1 outfit: formal black suit, white collared shirt, black necktie
Game 2 outfit: blue jacket, white collared shirt, black necktie
Appearance tags (optional): brown hair, stubble, gray eyes, 34 yo
After a failed test, I decided to try again. I think the previous test one was overbaked so I unlisted it.
This time I re-trained the LoRa on Civitai, using SDXL Pony and the default settings (just 418 steps). Used only 11 images with simplified tags.
Still not sure about the results this time, but they seem better at least. Will do more tests.
It's unfortunate, but I couldn't include any of his expressions in the dataset: the smile, the disapproval, the thinking pose etc. Maybe those can be reproduced in other ways.
After some initial tests:
The pencil drawing style seems to linger, so you may get better results by lowering the LoRa's strength and putting "pencil drawing" in the Negative prompt. If so, you can also use the appearance tags to compensate for the lowered strength. These tags were not included in the dataset so as to make them inherent to the LoRa, but feel free to use them.