đ simpletuner-lora
This is a LyCORIS adapter for text-to-image generation. It's derived from stabilityai/stable-diffusion-3.5-medium, offering enhanced capabilities for generating high - quality images.
đ Quick Start
This LyCORIS adapter is based on the stabilityai/stable-diffusion-3.5-medium model. The main validation prompt used during training was:
A photo-realistic image of a cat
⨠Features
- Based on Stable Diffusion 3.5 Medium: Leverages the power of the base model for high - quality image generation.
- LyCORIS Adapter: Enhances the model's performance with a LyCORIS adapter.
- Multiple Image Generation Modes: Supports text - to - image and image - to - image generation.
đĻ Installation
No specific installation steps are provided in the original README. If you want to use this adapter, you need to have the necessary dependencies installed, such as torch
, diffusers
, and lycoris
.
đģ Usage Examples
Basic Usage
import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights
def download_adapter(repo_id: str):
import os
from huggingface_hub import hf_hub_download
adapter_filename = "pytorch_lora_weights.safetensors"
cache_dir = os.environ.get('HF_PATH', os.path.expanduser('~/.cache/huggingface/hub/models'))
cleaned_adapter_path = repo_id.replace("/", "_").replace("\\", "_").replace(":", "_")
path_to_adapter = os.path.join(cache_dir, cleaned_adapter_path)
path_to_adapter_file = os.path.join(path_to_adapter, adapter_filename)
os.makedirs(path_to_adapter, exist_ok=True)
hf_hub_download(
repo_id=repo_id, filename=adapter_filename, local_dir=path_to_adapter
)
return path_to_adapter_file
model_id = 'stabilityai/stable-diffusion-3.5-medium'
adapter_repo_id = 'hmwhwm/simpletuner-lora'
adapter_filename = 'pytorch_lora_weights.safetensors'
adapter_file_path = download_adapter(repo_id=adapter_repo_id)
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16)
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
wrapper.merge_to()
prompt = "A photo-realistic image of a cat"
negative_prompt = 'blurry, cropped, ugly'
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
model_output = pipeline(
prompt=prompt,
negative_prompt=negative_prompt,
num_inference_steps=20,
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]
model_output.save("output.png", format="PNG")
đ Documentation
Validation settings
- CFG:
4.0
- CFG Rescale:
0.0
- Steps:
20
- Sampler:
FlowMatchEulerDiscreteScheduler
- Seed:
42
- Resolution:
1024x1024
- 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:
The text encoder was not trained. You may reuse the base model text encoder for inference.
Training settings
- Training epochs: 6
- Training steps: 3000
- Learning rate: 0.0001
- Learning rate schedule: polynomial
- Warmup steps: 100
- Max grad value: 2.0
- Effective batch size: 32
- Micro - batch size: 32
- Gradient accumulation steps: 1
- Number of GPUs: 1
- Gradient checkpointing: True
- Prediction type: flow_matching (extra parameters=['flow_schedule_auto_shift', 'shift=0.0', 'flow_use_uniform_schedule'])
- Optimizer: adamw_bf16
- Trainable parameter precision: Pure BF16
- Base model precision:
no_change
- Caption dropout probability: 0.1%
LyCORIS Config:
{
"algo": "lokr",
"multiplier": 1.0,
"full_matrix": true,
"linear_alpha": 1,
"factor": 16,
"apply_preset": {
"target_module": [
"Attention",
"FeedForward"
],
"module_algo_map": {
"Attention": {
"factor": 16
},
"FeedForward": {
"factor": 8
}
}
}
}
Datasets
pseudo - camera - 10k - sd3
- Repeats: 0
- Total number of images: 14102
- Total number of aspect buckets: 1
- Resolution: 1.048576 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square
- Used for regularisation data: No
đ§ Technical Details
This adapter is a LyCORIS adapter based on the Stable Diffusion 3.5 Medium model. It uses specific training settings and a LyCORIS configuration to enhance the model's performance. The validation settings are used to evaluate the model's performance during the validation phase.
đ License
The license is other.