metadata
license: cc-by-nc-4.0
Github repo: https://github.com/magic-research/piecewise-rectified-flow
Project page: https://piecewise-rectified-flow.github.io/
Demo:
import torch, torchvision
from diffusers import StableDiffusionPipeline, UNet2DConditionModel
from src.utils_perflow import merge_delta_weights_into_unet
from src.scheduler_perflow import PeRFlowScheduler
delta_weights = UNet2DConditionModel.from_pretrained("hansyan/piecewise-rectified-flow-delta-weights", torch_dtype=torch.float16, variant="v0-1",).state_dict()
pipe = StableDiffusionPipeline.from_pretrained("Lykon/dreamshaper-8", torch_dtype=torch.float16,)
pipe = merge_delta_weights_into_unet(pipe, delta_weights)
pipe.scheduler = PeRFlowScheduler.from_config(pipe.scheduler.config, prediction_type="epsilon", num_time_windows=4)
pipe.to("cuda", torch.float16)
prompts_list = ["A man with brown skin, a beard, and dark eyes", "A colorful bird standing on the tree, open beak",]
for i, prompt in enumerate(prompts_list):
generator = torch.Generator("cuda").manual_seed(1024)
prompt = "RAW photo, 8k uhd, dslr, high quality, film grain, highly detailed, masterpiece; " + prompt
neg_prompt = "distorted, blur, smooth, low-quality, warm, haze, over-saturated, high-contrast, out of focus, dark"
samples = pipe(
prompt = [prompt] * 8,
negative_prompt = [neg_prompt] * 8,
height = 512,
width = 512,
num_inference_steps = 8,
guidance_scale = 7.5,
generator = generator,
output_type = 'pt',
).images
torchvision.utils.save_image(torchvision.utils.make_grid(samples, nrow=4), f"tmp_{i}.png")