Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: zenless-lab/sdxl-anything-xl
|
| 3 |
+
tags:
|
| 4 |
+
- stable-diffusion-xl
|
| 5 |
+
- sdxl
|
| 6 |
+
- lora
|
| 7 |
+
- diffusers
|
| 8 |
+
- art
|
| 9 |
+
- style
|
| 10 |
+
license: other
|
| 11 |
+
language:
|
| 12 |
+
- en
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# LoRA Model
|
| 16 |
+
|
| 17 |
+
This is a LoRA (Low-Rank Adaptation) model trained on SDXL.
|
| 18 |
+
|
| 19 |
+
## Model Details
|
| 20 |
+
- **Base Model**: zenless-lab/sdxl-anything-xl
|
| 21 |
+
- **Training Method**: LoRA
|
| 22 |
+
- **LoRA Rank**: 64
|
| 23 |
+
- **Optimizer**: Prodigy
|
| 24 |
+
|
| 25 |
+
## Usage
|
| 26 |
+
|
| 27 |
+
```python
|
| 28 |
+
from diffusers import DiffusionPipeline
|
| 29 |
+
import torch
|
| 30 |
+
|
| 31 |
+
# Load pipeline
|
| 32 |
+
pipe = DiffusionPipeline.from_pretrained(
|
| 33 |
+
"zenless-lab/sdxl-anything-xl",
|
| 34 |
+
torch_dtype=torch.float16
|
| 35 |
+
).to("cuda")
|
| 36 |
+
|
| 37 |
+
# Load LoRA weights
|
| 38 |
+
pipe.load_lora_weights("dada22231/b426e506-75a2-4d9a-8952-22652d903752")
|
| 39 |
+
|
| 40 |
+
# Generate image
|
| 41 |
+
prompt = "your prompt in lora style"
|
| 42 |
+
image = pipe(prompt).images[0]
|
| 43 |
+
```
|
| 44 |
+
|
| 45 |
+
## Training Details
|
| 46 |
+
- Trained with AI Toolkit
|
| 47 |
+
- Mixed precision: bf16
|
| 48 |
+
- Batch size: 12
|