David Xie

I am faced with the similar problem. Maybe you can try to set the hyper-parameters as the original paper DDPM's. Through this way I get a little better performance, but...

https://github.com/lucidrains/denoising-diffusion-pytorch/issues/193 Here is my issue.

> 升级gradio包到4.39.0；使用下面的代码替换matte_anything.py中对应的部分 with gr.Blocks() as demo: gr.Markdown( """ # > > Matte Anything🐒 ! > """ > ) > with gr.Row(equal_height=True): #.style(equal_height=True): > with gr.Column(): > # input image >...

> > 升级gradio包到4.39.0；使用下面的代码替换matte_anything.py中对应的部分 with gr.Blocks() as demo: gr.Markdown( """ # > > Matte Anything🐒 ! > > """ > > ) > > with gr.Row(equal_height=True): #.style(equal_height=True): > > with gr.Column():...