|
--- |
|
license: mit |
|
tags: |
|
- pytorch |
|
- diffusers |
|
- unconditional-image-generation |
|
- diffusion-models-class |
|
--- |
|
|
|
This model is a diffusion model for unconditional image generation of churches ⛪️ finetuned on wikiart 🎨.<br> |
|
|
|
Pretrained model : google/ddpm-church-256<br> |
|
Dataset : huggan/wikiart<br> |
|
|
|
## Usage |
|
|
|
```python |
|
from diffusers import DDPMPipeline |
|
|
|
model_id = 'CCMat/ddpm-church-finetune-wikiart' |
|
|
|
# load model and scheduler |
|
pipeline = DDPMPipeline.from_pretrained(model_id) |
|
|
|
# run pipeline in inference (sample random noise and denoise) |
|
image = pipeline().images[0] |
|
|
|
# save image |
|
image.save("ddpm_church_wikiart.png") |
|
``` |
|
|
|
## Samples |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
|