bathroom_lora_1024

bathroom_lora_1024

This is a standard PEFT LoRA derived from stabilityai/stable-diffusion-3.5-large.

The main validation prompt used during training was:

b8m home bathroom interior design. The bright and clean bathroom area. The walls are finished with white tiles, and the ceiling is also neatly finished with white paint. The floor is finished with white tiles in a herringbone pattern, further emphasizing the bright feel of the entire space. The main color is white, silver is used as a secondary color. White makes the space look wide and clean, silver adds a sleek modern touch.

Validation settings

• CFG: 4.0
• CFG Rescale: 0.0
• Steps: 50
• Sampler: FlowMatchEulerDiscreteScheduler
• Seed: 42
• Resolutions: 1024x1024, 1024x896, 1280x768
• Skip-layer guidance: skip_guidance_layers=[7, 8, 9],

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery:

<Gallery />

The text encoder was not trained. You may reuse the base model text encoder for inference.

Training settings

• Training epochs: 20

• Training steps: 16000

• Learning rate: 3e-05

  • Learning rate schedule: polynomial
  • Warmup steps: 1000

• Max grad norm: 0.01

• Effective batch size: 1

  • Micro-batch size: 1
  • Gradient accumulation steps: 1
  • Number of GPUs: 1

• Gradient checkpointing: True

• Prediction type: flow-matching (extra parameters=['shift=1.0'])

• Optimizer: adamw_bf16

• Trainable parameter precision: Pure BF16

• Caption dropout probability: 30.0%

• LoRA Rank: 1024

• LoRA Alpha: None

• LoRA Dropout: 0.1

• LoRA initialisation style: default

Datasets

bathroom_lora_1024

• Repeats: 30
• Total number of images: 25
• Total number of aspect buckets: 1
• Resolution: 1.048576 megapixels
• Cropped: True
• Crop style: random
• Crop aspect: preserve
• Used for regularisation data: No

Inference

import torch
from diffusers import DiffusionPipeline

model_id = 'stabilityai/stable-diffusion-3.5-large'
adapter_id = 'daehuncho/bathroom_lora_1024'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
pipeline.load_lora_weights(adapter_id)

prompt = "b8m home bathroom interior design. The bright and clean bathroom area. The walls are finished with white tiles, and the ceiling is also neatly finished with white paint. The floor is finished with white tiles in a herringbone pattern, further emphasizing the bright feel of the entire space. The main color is white, silver is used as a secondary color. White makes the space look wide and clean, silver adds a sleek modern touch."
negative_prompt = 'blurry, cropped, ugly'

## Optional: quantise the model to save on vram.
## Note: The model was quantised during training, and so it is recommended to do the same during inference time.
from optimum.quanto import quantize, freeze, qint8
quantize(pipeline.transformer, weights=qint8)
freeze(pipeline.transformer)
    
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
image = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=50,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(42),
    width=1024,
    height=1024,
    guidance_scale=4.0,
    skip_guidance_layers=[7, 8, 9],
).images[0]
image.save("output.png", format="PNG")