Text-to-Image
Diffusers
PyTorch
StableDiffusionPipeline
stable-diffusion
diffusion-models-class
dreambooth-hackathon
science
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("baruga/mandelbrot-fractals", dtype=torch.bfloat16, device_map="cuda")
prompt = "top rated photo of mafra fractal in the shape of seashells."
image = pipe(prompt).images[0]Description
This is a Stable Diffusion model fine-tuned on Mandelbrot fractal images for the DreamBooth Hackathon 🔥 science theme. To participate or learn more, visit this page.
To generate Mandelbrot fractals, use a photo of mafra fractal in the shape of [your choice] or experiment with other variations. CFG scale seems to be the best around 8-9. Additional modifiers and negative prompts may also improve results.
Examples
a photo of mafra fractal in the shape of a squid.
a photo of mafra fractal in the shape of seashells.
a photo of mafra fractal in the shape of jungle foliage.
a photo of mafra fractal in the shape of a beautiful flower.
Usage
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained('baruga/mandelbrot-fractals')
image = pipeline().images[0]
image
- Downloads last month
- 3