dore-engraving-5e-4-ss3.0

This is a LyCORIS adapter derived from black-forest-labs/FLUX.1-dev.

No validation prompt was used during training.

None

Validation settings

  • CFG: 2.5
  • CFG Rescale: 0.0
  • Steps: 20
  • Sampler: FlowMatchEulerDiscreteScheduler
  • Seed: 42
  • Resolution: 1024x1024
  • Skip-layer guidance:

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

You can find some example images in the following gallery:

Prompt
unconditional (blank prompt)
Negative Prompt
blurry, cropped, ugly
Prompt
a hipster man with a beard, building a chair in the style of a d0r3 engraving.
Negative Prompt
blurry, cropped, ugly
Prompt
a hamster in the style of a d0r3 engraving.
Negative Prompt
blurry, cropped, ugly
Prompt
a range rover on the beach in the style of a d0r3 engraving.
Negative Prompt
blurry, cropped, ugly
Prompt
a Coca-Cola can product shot in the style of a d0r3 engraving.
Negative Prompt
blurry, cropped, ugly
Prompt
a man holding a sign that says, 'this is a sign
Negative Prompt
blurry, cropped, ugly
Prompt
a pig, in a post apocalyptic world, with a shotgun, in a leather jacket, in a desert, with a motorcycle
Negative Prompt
blurry, cropped, ugly
Prompt
woman holding a sign that says 'I LOVE PROMPTS!' in the style of a d0r3 engraving
Negative Prompt
blurry, cropped, ugly
Prompt
two men in robes with laurel wreaths in a haunted forest with gnarled branches in the style of a d0r3 engraving.
Negative Prompt
blurry, cropped, ugly
Prompt
A person kneels on the ground with a staff. Three figures with wings stand elevated on the left side. Background shows a cloudy sky and hilly terrain in the style of a d0r3 engraving.
Negative Prompt
blurry, cropped, ugly

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

Training settings

  • Training epochs: 3
  • Training steps: 2900
  • Learning rate: 0.0005
    • Learning rate schedule: polynomial
    • Warmup steps: 100
  • Max grad norm: 0.1
  • Effective batch size: 3
    • Micro-batch size: 3
    • Gradient accumulation steps: 1
    • Number of GPUs: 1
  • Gradient checkpointing: True
  • Prediction type: flow-matching (extra parameters=['shift=3.0', 'flux_guidance_mode=constant', 'flux_guidance_value=1.0', 'flow_matching_loss=compatible'])
  • Optimizer: adamw_bf16
  • Trainable parameter precision: Pure BF16
  • Caption dropout probability: 10.0%

LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1.0,
    "linear_dim": 10000,
    "linear_alpha": 1,
    "factor": 16,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 16
            },
            "FeedForward": {
                "factor": 8
            }
        }
    }
}

Datasets

cassatt-combined-512

  • Repeats: 11
  • Total number of images: 60
  • Total number of aspect buckets: 1
  • Resolution: 0.262144 megapixels
  • Cropped: False
  • Crop style: None
  • Crop aspect: None
  • Used for regularisation data: No

cassatt-combined-768

  • Repeats: 11
  • Total number of images: 59
  • Total number of aspect buckets: 1
  • Resolution: 0.589824 megapixels
  • Cropped: False
  • Crop style: None
  • Crop aspect: None
  • Used for regularisation data: No

cassatt-combined-1024

  • Repeats: 5
  • Total number of images: 59
  • Total number of aspect buckets: 6
  • Resolution: 1.048576 megapixels
  • Cropped: False
  • Crop style: None
  • Crop aspect: None
  • Used for regularisation data: No

cassatt-oil-1536

  • Repeats: 2
  • Total number of images: 58
  • Total number of aspect buckets: 1
  • Resolution: 2.359296 megapixels
  • Cropped: False
  • Crop style: None
  • Crop aspect: None
  • Used for regularisation data: No

Inference

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 = 'black-forest-labs/FLUX.1-dev'
adapter_repo_id = 'mipat12/dore-engraving-5e-4-ss3.0'
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) # loading directly in bf16
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
wrapper.merge_to()

prompt = "An astronaut is riding a horse through the jungles of Thailand."


## 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,
    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=2.5,
).images[0]
image.save("output.png", format="PNG")

Exponential Moving Average (EMA)

SimpleTuner generates a safetensors variant of the EMA weights and a pt file.

The safetensors file is intended to be used for inference, and the pt file is for continuing finetuning.

The EMA model may provide a more well-rounded result, but typically will feel undertrained compared to the full model as it is a running decayed average of the model weights.

Downloads last month
253
Inference API
Examples

Model tree for mipat12/dore-engraving-5e-4-ss3.0

Adapter
(12097)
this model