CoolFace
Modelpublic

uxoah/fantasy_dev_lokr_adamw_2e-4_ema_v2

sourceHugging Faceotherupdated 2y agoView on Hugging Face
0likes7downloads
Model Card

fantasydevlokradamw2e-4emav2

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

The main validation prompt used during training was:

A girl sitting on a box

Validation settings

  • —CFG: 3.0
  • —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:

<Gallery />

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

Training settings

  • —Training epochs: 70
  • —Training steps: 4000
  • —Learning rate: 0.0005
  • —Learning rate schedule: polynomial
  • —Warmup steps: 200
  • —Max grad norm: 0.01
  • —Effective batch size: 2
  • —Micro-batch size: 2
  • —Gradient accumulation steps: 1
  • —Number of GPUs: 1
  • —Gradient checkpointing: True
  • —Prediction type: flow-matching (extra parameters=['shift=3', 'fluxguidancemode=constant', 'fluxguidancevalue=1.0', 'flowmatchingloss=compatible'])
  • —Optimizer: adamw_bf16
  • —Trainable parameter precision: Pure BF16
  • —Caption dropout probability: 0.0%

LyCORIS Config:

json
{
    "algo": "lokr",
    "bypass_mode": true,
    "multiplier": 1.0,
    "full_matrix": true,
    "linear_dim": 10000,
    "linear_alpha": 1,
    "factor": 10,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 10
            },
            "FeedForward": {
                "factor": 4
            }
        }
    }
}

Datasets

domonikmayerartsource2ST-1024

  • —Repeats: 0
  • —Total number of images: 33
  • —Total number of aspect buckets: 1
  • —Resolution: 1.048576 megapixels
  • —Cropped: False
  • —Crop style: None
  • —Crop aspect: None
  • —Used for regularisation data: No

domonikmayerartsource2ST-768

  • —Repeats: 0
  • —Total number of images: 39
  • —Total number of aspect buckets: 2
  • —Resolution: 0.589824 megapixels
  • —Cropped: False
  • —Crop style: None
  • —Crop aspect: None
  • —Used for regularisation data: No

domonikmayerartsource2ST-512

  • —Repeats: 0
  • —Total number of images: 39
  • —Total number of aspect buckets: 2
  • —Resolution: 0.262144 megapixels
  • —Cropped: False
  • —Crop style: None
  • —Crop aspect: None
  • —Used for regularisation data: No

Inference

python
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 = 'uxoah/fantasy_dev_lokr_adamw_2e-4_ema_v2'
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 = "A girl sitting on a box"


## 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=3.0,
).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.