LoRA trained on 35 publicly available images of Yvonne Strahovski.
Trained on base SD 1.5, should work with most models.
Use ADetailer if details are lost and strength of 0.74 should work on most prompts. Use lower strength for other models.
(25, 14 images) - (24, 16 repeats). Picked the 7th epoch.
Consider leaving a review if you like the model. All feedback is appreciated.
Description
FAQ
Comments (5)
i love it. this is the hands down one of the best loras out there. you could easily say that some of those images are real.
may i ask the source, resolution of training images? did you clear the background? blurred it? maybe share training template? did you resize training images?
Thanks
My entire training workflow is described here in detail.
Celebrity LoRA Training Guide (Consolidated) - SD 1.5 | Civitai
I did not remove/blur the background for the training data.
@__2_ thank you for your response. Looking forward to your next lora :)
@Khser Any suggestions?
@__2_ not really, looking at your loras and they are very very very good. i am kinda thinking of stealing your ideas and train sdxl versions of them :D (i am still learning so trying to come up with ideas)