---
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/WangXin93/three-kingdoms-stable-diffusion) with this [dataset](https://huggingface.co/datasets/wx44wx/three-kingdoms-blip-captions).

> a man in armor
![image.png](https://github.com/WangXin93/three-kingdoms-stable-diffusion/raw/main/assets/a-man-in-armor.png)

> a women in red dress
![image.png](https://github.com/WangXin93/three-kingdoms-stable-diffusion/raw/main/assets/a-women-in-red-dress.png)

> a women in armor
 ![image.png](https://github.com/WangXin93/three-kingdoms-stable-diffusion/raw/main/assets/a-women-in-armor.png)

## Usage

```bash
!pip install diffusers==0.19.3
!pip install transformers scipy ftfy
```

```python
 python scripts/txt2img.py \
    --prompt 'a man in armor' \
    --outdir 'outputs/generated_three_kingdoms_officers' \
    --H 512 --W 512 \
    --n_samples 4 \
    --config 'configs/stable-diffusion/three-kingdoms.yaml' \
    --ckpt ema-only-epoch=000144.ckpt
```

## 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/LambdaLabsML/examples/tree/main/stable-diffusion-finetuning) for the code.