File size: 2,570 Bytes
09b2f5a 590eb27 09b2f5a 590eb27 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 |
---
license: mit
language:
- en
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
datasets:
- wx44wx/three-kingdoms-blip-captions
---
__Stable Diffusion fine tuned on [Romance of the Three Kingdoms XI: Officer Portraits](https://kongming.net/11/portraits/).__
Put in a text prompt and generate your own Officier in Three Kingdoms.
trained using this [script](https://github.com/LambdaLabsML/examples/tree/main/stable-diffusion-finetuning) with this [dataset](https://huggingface.co/datasets/wx44wx/three-kingdoms-blip-captions).
> a man in armor
> a women in red dress
> a women in armor
## Usage
```bash
!pip install diffusers==0.19.3
!pip install transformers scipy ftfy
```
```python
import torch
from diffusers import StableDiffusionPipeline
from torch import autocast
pipe = StableDiffusionPipeline.from_pretrained("wx44wx/sd-three-kingdoms-diffusers", torch_dtype=torch.float16)
pipe = pipe.to("cuda")
prompt = "a man in armor"
scale = 3
n_samples = 4
# Sometimes the nsfw checker is confused by the Pokémon images, you can disable
# it at your own risk here
disable_safety = False
if disable_safety:
def null_safety(images, **kwargs):
return images, False
pipe.safety_checker = null_safety
with autocast("cuda"):
images = pipe(n_samples*[prompt], guidance_scale=scale).images
for idx, im in enumerate(images):
im.save(f"{idx:06}.png")
```
## Model description
Trained on [BLIP captioned Three Kingdoms Officers images](https://huggingface.co/datasets/wx44wx/three-kingdoms-blip-captions) using 1xA6000 GPUs for around 16,000 steps.
## Links
- [Lambda Diffusers](https://github.com/LambdaLabsML/lambda-diffusers)
- [Captioned Three Kingdoms dataset](https://huggingface.co/datasets/wx44wx/three-kingdoms-blip-captions)
- [Model weights in Diffusers format](https://huggingface.co/wx44wx/sd-three-kingdoms-diffusers)
- [Original model weights](https://huggingface.co/wx44wx/three-kingdoms-stable-diffusion)
- [Training code](https://github.com/justinpinkney/stable-diffusion)
Trained by [Xin Wang](wangxin93.github.io). Thanks [kongming.net](kongming.net) for their archived images and [justinpinkney](https://github.com/justinpinkney/stable-diffusion) for the code. |