README.md

---
language: zh
license: creativeml-openrail-m

tags:

- diffusion
- zh
- Chinese
---



# Midu-Stable-Diffusion-2-Chinese-Style-v0.1



## Brief Introduction

| ![cyberpunk](examples/cyberpunk.jpeg) | ![shiba](examples/shiba.jpeg) | ![ds](examples/ds.jpeg)         |
| ------------------------------------- | ----------------------------- | ------------------------------- |
| ![waitan](examples/waitan.jpeg)       | ![gf](examples/gf.jpeg)       | ![ssh](examples/ssh.jpeg)       |
| ![cat](examples/cat.jpeg)             | ![robot](examples/robot.jpeg) | ![castle](examples/castle.jpeg) |

大概是Huggingface 🤗社区首个开源的Stable diffusion 2 中文模型。该模型基于stable diffusion V2.1模型，在约500万条的中国风格特挑中文数据上进行微调，数据来源于多个开源数据集如[LAION-5B](https://laion.ai/blog/laion-5b/),  [Noah-Wukong](https://wukong-dataset.github.io/wukong-dataset/), [Zero](https://zero.so.com/)和一些网络数据。

Probably the first open sourced Chinese Stable Diffusion 2 model in Huggingface🤗 community. This model is finetuned based on stable diffusion V2.1 with 5M chinese style filtered data. Dataset is composed of several different chinese open source dataset such as [LAION-5B](https://laion.ai/blog/laion-5b/),  [Noah-Wukong](https://wukong-dataset.github.io/wukong-dataset/), [Zero](https://zero.so.com/) and some web data.



## Model Details

### 文本编码器/Text Encoder

文本编码器使用冻结参数的[lyua1225/clip-huge-zh-75k-steps-bs4096](https://huggingface.co/lyua1225/clip-huge-zh-75k-steps-bs4096)。

Text encoder is frozen [lyua1225/clip-huge-zh-75k-steps-bs4096](https://huggingface.co/lyua1225/clip-huge-zh-75k-steps-bs4096) .

### Unet

在特挑的500万中文数据集上训练了150K steps，使用指数移动平均值(EMA)做原绘画能力保留，使模型能够在中文风格和原绘画能力之间获得权衡。

Training on 5M chinese style filtered data for 150k steps. Exponential moving average(EMA) is applied to keep the original Stable Diffusion 2 drawing capability and reach a balance between chinese style and original drawing capability.


## Usage

因为使用了customed tokenizer, 所以需要优先加载一下tokenizer

```py
# !pip install git+https://github.com/huggingface/accelerate
import torch
from diffusers import StableDiffusionPipeline
torch.backends.cudnn.benchmark = True
pipe = StableDiffusionPipeline.from_pretrained("IDEA-CCNL/Taiyi-Stable-Diffusion-1B-Chinese-v0.1", torch_dtype=torch.float16)
pipe.to('cuda')

prompt = '飞流直下三千尺，油画'
image = pipe(prompt, guidance_scale=7.5).images[0]  
image.save("飞流.png")
```