fix link and add usage code
Browse files
README.md
CHANGED
@@ -33,6 +33,8 @@ widget:
|
|
33 |
base_model: stabilityai/stable-diffusion-xl-base-1.0
|
34 |
instance_prompt: factorio
|
35 |
license: cc0-1.0
|
|
|
|
|
36 |
---
|
37 |
# Factorio
|
38 |
|
@@ -40,15 +42,42 @@ license: cc0-1.0
|
|
40 |
|
41 |
## Model description
|
42 |
|
43 |
-
Create abstract art in the style of artfully visualized Factorio blueprints. This Lora is trained using this dataset: https
|
44 |
|
45 |
## Trigger words
|
46 |
|
47 |
You should use `factorio` to trigger the image generation.
|
48 |
|
49 |
-
|
50 |
## Download model
|
51 |
|
52 |
Weights for this model are available in Safetensors format.
|
53 |
|
54 |
[Download](/piebro/factorio-blueprint-visualizations-sdxl-lora/tree/main) them in the Files & versions tab.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
base_model: stabilityai/stable-diffusion-xl-base-1.0
|
34 |
instance_prompt: factorio
|
35 |
license: cc0-1.0
|
36 |
+
datasets:
|
37 |
+
- piebro/factorio-blueprint-visualizations
|
38 |
---
|
39 |
# Factorio
|
40 |
|
|
|
42 |
|
43 |
## Model description
|
44 |
|
45 |
+
Create abstract art in the style of artfully visualized Factorio blueprints. This Lora is trained using this dataset: https://huggingface.co/datasets/piebro/factorio-blueprint-visualizations.
|
46 |
|
47 |
## Trigger words
|
48 |
|
49 |
You should use `factorio` to trigger the image generation.
|
50 |
|
|
|
51 |
## Download model
|
52 |
|
53 |
Weights for this model are available in Safetensors format.
|
54 |
|
55 |
[Download](/piebro/factorio-blueprint-visualizations-sdxl-lora/tree/main) them in the Files & versions tab.
|
56 |
+
|
57 |
+
## Usage
|
58 |
+
|
59 |
+
```python
|
60 |
+
import torch
|
61 |
+
from diffusers import DiffusionPipeline, AutoencoderKL
|
62 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
63 |
+
vae = AutoencoderKL.from_pretrained('madebyollin/sdxl-vae-fp16-fix', torch_dtype=torch.float16)
|
64 |
+
pipe = DiffusionPipeline.from_pretrained(
|
65 |
+
"stabilityai/stable-diffusion-xl-base-1.0",
|
66 |
+
vae=vae, torch_dtype=torch.float16, variant="fp16",
|
67 |
+
use_safetensors=True
|
68 |
+
)
|
69 |
+
pipe.to(device)
|
70 |
+
# This is where you load your trained weights
|
71 |
+
specific_safetensors = "pytorch_lora_weights.safetensors"
|
72 |
+
lora_scale = 0.9
|
73 |
+
pipe.load_lora_weights(
|
74 |
+
'piebro/factorio-blueprint-visualizations-sdxl-lora',
|
75 |
+
weight_name = specific_safetensors,
|
76 |
+
)
|
77 |
+
prompt = "pastel drawing of factorio"
|
78 |
+
image = pipe(
|
79 |
+
prompt=prompt,
|
80 |
+
num_inference_steps=50,
|
81 |
+
cross_attention_kwargs={"scale": lora_scale}
|
82 |
+
).images[0]
|
83 |
+
```
|