Shivam Dreambooth>LORA of British singer/songwriter Kate Bush, using 165 ref images and 1980 (real) reg images, for 20,000 steps. The period covered is around 1978-82.

This is a tricky model to use, and among the least 'one-click' of my models. In general, keep the sampling steps no higher than about 60, ideally on the good Karras interpreters, and keep working the picture until it assembles. You can move the steps up as you go in for detail later.

I generally use these models in complex workflows, inpainting faces after initial T2Is, and using ControlNet extensively. So if you're hoping for one-click prompt magic, my models aren't data-curated with this in mind, but rather as tools for traditional workflows that use Photoshop and other older methods.

This model is very likely to produce NSFW and nude renderings unless counter-prompted.