|
|
--- |
|
|
license: apache-2.0 |
|
|
datasets: |
|
|
- Rapidata/Face_Generation_Benchmark |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- stabilityai/stable-diffusion-xl-base-1.0 |
|
|
pipeline_tag: text-to-image |
|
|
library_name: diffusers |
|
|
tags: |
|
|
- text-to-image |
|
|
- lora |
|
|
- diffusers |
|
|
--- |
|
|
## Model description |
|
|
|
|
|
# Portrait-generator |
|
|
|
|
|
**Portrait-generator** is a LoRA fine-tuned adapter for the [stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) model, trained on the [Rapidata/Face_Generation_Benchmark dataset](https://huggingface.co/datasets/Rapidata/Face_Generation_Benchmark). It specializes in generating face images. |
|
|
|
|
|
--- |
|
|
|
|
|
## Usage |
|
|
|
|
|
1. **Install dependencies** (if not installed): |
|
|
|
|
|
```bash |
|
|
pip install diffusers transformers accelerate safetensor torch |
|
|
``` |
|
|
|
|
|
2. **Load stable-diffusion-xl-base-1.0 and LoRA adapter**: |
|
|
|
|
|
```python |
|
|
from diffusers import DiffusionPipeline |
|
|
import torch |
|
|
|
|
|
pipe = DiffusionPipeline.from_pretrained( |
|
|
"stabilityai/stable-diffusion-xl-base-1.0", |
|
|
torch_dtype=torch.float16, |
|
|
).to("cuda") |
|
|
|
|
|
pipe.load_lora_weights("SkyAsl/Portrait-generator") |
|
|
pipe.fuse_lora() |
|
|
|
|
|
image = pipe("a high-quality portrait of a woman").images[0] |
|
|
image.save("test.png") |
|
|
``` |
|
|
|
|
|
--- |
|
|
|
|
|
# Metrics |
|
|
|
|
|
- Average CLIPScore: 16.6980 |
|
|
|
|
|
--- |
