growwithdaisy/crrllcrrllxovrtn_styles_20241112_172347

This is a LyCORIS adapter derived from FLUX.1-dev.

The main validation prompt used during training was:

a photo of a daisy

Validation settings

CFG: 3.5
CFG Rescale: 0.0
Steps: 20
Sampler: None
Seed: 69
Resolution: 1024x1024

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
jmmymrbl style

Negative Prompt
blurry, cropped, ugly

Prompt
jmmymrbl cutup style

Negative Prompt
blurry, cropped, ugly

Prompt
jmmymrbl earthly delights style

Negative Prompt
blurry, cropped, ugly

Prompt
jmmymrbl dunes style

Negative Prompt
blurry, cropped, ugly

Prompt
a photo of a daisy

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: 51
Training steps: 2500
Learning rate: 0.0002
Max grad norm: 2.0
Effective batch size: 8
- Micro-batch size: 1
- Gradient accumulation steps: 1
- Number of GPUs: 8
Prediction type: flow-matching (extra parameters=['shift=3', 'flux_guidance_value=1.0'])
Rescaled betas zero SNR: False
Optimizer: optimi-stableadamwweight_decay=1e-3
Precision: Pure BF16
Quantised: No
Xformers: Not used
LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1,
    "linear_dim": 1000000,
    "linear_alpha": 1,
    "factor": 16,
    "init_lokr_norm": 0.001,
    "apply_preset": {
        "target_module": [
            "FluxTransformerBlock",
            "FluxSingleTransformerBlock"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 16
            },
            "FeedForward": {
                "factor": 8
            }
        }
    }
}

Datasets

jmmymrbl_general_style-512

Repeats: 0
Total number of images: ~56
Total number of aspect buckets: 4
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_general_style-768

Repeats: 0
Total number of images: ~56
Total number of aspect buckets: 5
Resolution: 0.589824 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_cutup_style-512

Repeats: 0
Total number of images: ~32
Total number of aspect buckets: 3
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_cutup_style-768

Repeats: 0
Total number of images: ~32
Total number of aspect buckets: 2
Resolution: 0.589824 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_earthly_delights_style-512

Repeats: 0
Total number of images: ~32
Total number of aspect buckets: 4
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_earthly_delights_style-768

Repeats: 0
Total number of images: ~40
Total number of aspect buckets: 5
Resolution: 0.589824 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_earthly_delights_style-1024

Repeats: 0
Total number of images: ~56
Total number of aspect buckets: 7
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_dunes_style-512

Repeats: 0
Total number of images: ~48
Total number of aspect buckets: 4
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

jmmymrbl_dunes_style-768

Repeats: 0
Total number of images: ~40
Total number of aspect buckets: 3
Resolution: 0.589824 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 = 'FLUX.1-dev'
adapter_repo_id = 'playerzer0x/growwithdaisy/crrllcrrllxovrtn_styles_20241112_172347'
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 photo of a daisy"


## Optional: quantise the model to save on vram.
## Note: The model was not quantised during training, so it is not necessary to quantise it 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(1641421826),
    width=1024,
    height=1024,
    guidance_scale=3.5,
).images[0]
image.save("output.png", format="PNG")