Dushwe/mineral-colour

mineral-colour on Stable Diffusion via Dreambooth

token

mineral-colour

Here are the images used for training this concept:

inference

from torch import autocast
from diffusers import StableDiffusionPipeline
import torch
import diffusers
from PIL import Image
def image_grid(imgs, rows, cols):
    assert len(imgs) == rows*cols
    w, h = imgs[0].size
    grid = Image.new('RGB', size=(cols*w, rows*h))
    grid_w, grid_h = grid.size                    
    for i, img in enumerate(imgs):
        grid.paste(img, box=(i%cols*w, i//cols*h))
    return grid
pipe = StableDiffusionPipeline.from_pretrained("Dushwe/mineral-colour").to("cuda")
prompt = 'A little girl in china chic hanfu walks in the forest, mineral-colour'
images = pipe(prompt, num_images_per_prompt=1, num_inference_steps=50, guidance_scale=7.5,torch_dtype=torch.cuda.HalfTensor).images
grid = image_grid(images, 1, 1)
grid

generate samples

Chinese palace, 4k resolution, mineral-colour

beginning of autumn, autumn, forests, scenery, background, landscape, woodland, trees,mineral-colour

You run your new concept via diffusers Colab Notebook for Inference. Don't forget to use the concept prompts!