metadata

license: other
base_model: black-forest-labs/FLUX.1-dev
tags:
  - flux
  - flux-diffusers
  - text-to-image
  - diffusers
  - simpletuner
  - not-for-all-audiences
  - lora
  - template:sd-lora
  - lycoris
inference: true
widget:
  - text: unconditional (blank prompt)
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_0_0.png
  - text: >-
      xnywng woman, The image shows a young woman standing in front of a white
      brick wall with her arms stretched out to the sides. She is wearing a pink
      jacket, a white crop top, and a pair of pink pants. She has a black
      baseball cap on her head and is smiling at the camera. On the right side
      of the image, there are several green plastic
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_1_0.png
  - text: dsndsn clock, yellow red and blue
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_2_0.png
  - text: >-
      glssr skywash, two frames with out-of-focus background, tubes and
      applicator being held by someone
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_3_0.png
  - text: glssr mini beauty bag, red, on a glossier bag, on a plaster background
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_4_0.png
  - text: a photo of a daisy
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_5_0.png

growwithdaisy/glssrxdsndsn_flat_20241212_212607

This is a LyCORIS adapter derived from black-forest-labs/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: FlowMatchEulerDiscreteScheduler
Seed: 69
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
xnywng woman, The image shows a young woman standing in front of a white brick wall with her arms stretched out to the sides. She is wearing a pink jacket, a white crop top, and a pair of pink pants. She has a black baseball cap on her head and is smiling at the camera. On the right side of the image, there are several green plastic

Negative Prompt
blurry, cropped, ugly

Prompt
dsndsn clock, yellow red and blue

Negative Prompt
blurry, cropped, ugly

Prompt
glssr skywash, two frames with out-of-focus background, tubes and applicator being held by someone

Negative Prompt
blurry, cropped, ugly

Prompt
glssr mini beauty bag, red, on a glossier bag, on a plaster background

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: 56
Training steps: 5000
Learning rate: 0.0002
- Learning rate schedule: constant
- Warmup steps: 0
Max grad norm: 2.0
Effective batch size: 16
- Micro-batch size: 2
- Gradient accumulation steps: 1
- Number of GPUs: 8
Gradient checkpointing: True
Prediction type: flow-matching (extra parameters=['shift=3', 'flux_guidance_mode=constant', 'flux_guidance_value=1.0', 'flow_matching_loss=compatible'])
Optimizer: optimi-stableadamwweight_decay=1e-3
Trainable parameter precision: Pure BF16
Caption dropout probability: 5.0%

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

glssrxdsndsn_flat-512

Repeats: 0
Total number of images: ~336
Total number of aspect buckets: 2
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

glssrxdsndsn_flat-768

Repeats: 0
Total number of images: ~296
Total number of aspect buckets: 5
Resolution: 0.589824 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

glssrxdsndsn_flat-1024

Repeats: 1
Total number of images: ~248
Total number of aspect buckets: 12
Resolution: 1.048576 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 = 'playerzer0x/growwithdaisy/glssrxdsndsn_flat_20241212_212607'
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(69),
    width=1024,
    height=1024,
    guidance_scale=3.5,
).images[0]
image.save("output.png", format="PNG")