Add Stable Diffusion XL Long Weighted Prompt Pipeline

[co63oc]

#413

paddle

torch

paddle, torch生成图一致

torch 代码

from diffusers import DiffusionPipeline
import torch

pipe = DiffusionPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0"
    , torch_dtype       = torch.float16
    , use_safetensors   = True
    , variant           = "fp16"
    , custom_pipeline   = "lpw_stable_diffusion_xl",
)

prompt = "photo of a cute (white) cat running on the grass"*20
prompt2 = "chasing (birds:1.5)"*20
prompt = f"{prompt},{prompt2}"
neg_prompt = "blur, low quality, carton, animate"

pipe.to("cuda")
images = pipe(
    prompt                  = prompt
    , negative_prompt       = neg_prompt
    , generator=torch.Generator("cuda").manual_seed(452)
).images[0]
images.save("torch_out.png")

[paddle-bot]

Thanks for your contribution!

[westfish]

抱歉忘了强调版本了，需要参考diffusers 0.24.0的相关代码来实现，当前develop与diffusers 0.24.0对齐

[co63oc]

抱歉忘了强调版本了，需要参考diffusers 0.24.0的相关代码来实现，当前develop与diffusers 0.24.0对齐

使用的是对齐diffusers 0.24.0的develop版本

[westfish]

diffusers 0.24.0貌似不需要用到StableDiffusionMixin

[co63oc]

diffusers 0.24.0貌似不需要用到StableDiffusionMixin

已修改取消 StableDiffusionMixin

[westfish]

LGTM

[co63oc]

@westfish CI已完成

[co63oc]

@westfish 这里怎么不是像Paddle那样，Paddle不用更新分支