Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
IDEA-CCNL
/
Taiyi-Stable-Diffusion-1B-Chinese-EN-v0.1
like
105
Follow
Fengshenbang-LM
241
Text-to-Image
Diffusers
Chinese
StableDiffusionPipeline
stable-diffusion
stable diffusion chinese
stable-diffusion-diffusers
Chinese
Inference Endpoints
arxiv:
2112.10752
arxiv:
2209.02970
License:
creativeml-openrail-m
Model card
Files
Files and versions
Community
11
Train
Deploy
Use this model
感谢您的工作,这里有两个关于训练阶段的问题想请教下
#9
by
wqf321
- opened
Feb 6, 2023
Discussion
wqf321
Feb 6, 2023
我观察到您发布了两版中文diffusion模型,并且采取了不同的训练策略
为什么使用two stage的训练方法,即在第二阶段中将全部模型解冻,一起训练text encoder和diffusion model的方式可以让模型具备双语能力呢?我看数据集并没有变化
在中文图文生成任务上,使用单阶段的训练策略,即全程只训练text encoder,效果会显著优于两阶段的训练策略吗?
Edit
Preview
Upload images, audio, and videos by dragging in the text input, pasting, or
clicking here
.
Tap or paste here to upload images
Comment
·
Sign up
or
log in
to comment