---
tags:
- discodiffusion
- guideddiffusion
thumbnail: https://de.gravatar.com/userimage/52045156/8ab369c1d246e65bda88813ce7c4cb81.jpeg
datasets:
- wikiart

---


# Ukiyo-e Diffusion

If you make something using these models, you're welcome to mention me [@thegenerativegeneration](https://www.instagram.com/thegenerativegeneration/)


Named by dataset used. Current and best version is [models/ukiyoe-all/v1/ema_0.9999_056000.pt](models/ukiyoe-all/v1/ema_0.9999_056000.pt)

# Current Plans

* clean dataset
  * remove borders
  * remove some of the samples with text in them

# Models

## Ukiyo-e-all

### v1

[models/ukiyoe-all/v1/ema_0.9999_056000.pt](models/ukiyoe-all/v1/ema_0.9999_056000.pt)

Model configuration is:

```python
model_config = {
    'attention_resolutions': '32, 16, 8',
    'class_cond': False,
    'image_size': 256,
    'learn_sigma': True,
    'rescale_timesteps': True,
    'noise_schedule': 'linear',
    'num_channels': 128,
    'num_heads': 4,
    'num_res_blocks': 2,
    'resblock_updown': True,
    'use_checkpoint': True,
    'use_fp16': True,
    'use_scale_shift_norm': True,
}
```
#### Tips

- Results closest to original training data are achieved by turning off the secondary model in Disco Diffusion.
- Turning secondary model on can lead to very creative results
- It is not necessary to specify Ukiyo-e as artstyle to get ukiyo-e-like images.

#### Examples

If you make something nice using these models, I would like to link your image.

##### Secondary Off

![](models/ukiyoe-all/v1/images/secondary_off_3.png)
![](models/ukiyoe-all/v1/images/secondary_off_0.png)
![](models/ukiyoe-all/v1/images/secondary_off_1.png)
![](models/ukiyoe-all/v1/images/secondary_off_2.png)

##### Secondary On

![](models/ukiyoe-all/v1/images/secondary_on_0.png)
![](models/ukiyoe-all/v1/images/secondary_on_1.png)
![](models/ukiyoe-all/v1/images/secondary_on_2.png)


#### About
Trained from scratch on a ~170000 images corpus of [ukiyo-e.org](https://ukiyo-e.org) filtered by [colorfulness](https://pyimagesearch.com/2017/06/05/computing-image-colorfulness-with-opencv-and-python/
) >= 5.




## (Deprecated) Ukiyo-e-few

[models/ukiyoe-few/v1/ukiyoe_diffusion_256_022000.pt](models/ukiyoe-few/v1/ukiyoe_diffusion_256_022000.pt)

Finetuned on 5224 images from Wikiart (1168) and ? (). 


Model configuration is

```python
model_config = {
	'attention_resolutions': '16',
	'class_cond': False,
	'diffusion_steps': 1000,
	'rescale_timesteps': True,
	'timestep_respacing': 'ddim100',
	'image_size': 256,
	'learn_sigma': True,
	'noise_schedule': 'linear',
	'num_channels': 128,
	'num_heads': 1,
	'num_res_blocks': 2,
	'use_checkpoint': True,
	'use_scale_shift_norm': False
}
```

Trained using a fork of [guided-diffusion-sxela](https://github.com/thegenerativegeneration/guided-diffusion-sxela). Added random crop which did not lead to good results.