playerzer0x/barcelona_knoll_chair_20241025_184822

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

The main validation prompt used during training was:

photo of a woman sitting on a bscspc barcelona knoll chair

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
bscspc barcelona knoll chair, white background, front view, facing right

Negative Prompt
blurry, cropped, ugly

Prompt
bscspc barcelona knoll chair, white background, front view, facing left

Negative Prompt
blurry, cropped, ugly

Prompt
bscspc barcelona knoll chair, closeup, white background, front view, facing left

Negative Prompt
blurry, cropped, ugly

Prompt
photo of a woman sitting on a bscspc barcelona knoll chair

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: 42
Training steps: 300
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 (flux 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

barcelona_knoll_chair-512

Repeats: 0
Total number of images: ~24
Total number of aspect buckets: 2
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

barcelona_knoll_chair-768

Repeats: 0
Total number of images: ~16
Total number of aspect buckets: 1
Resolution: 0.589824 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

barcelona_knoll_chair-1024

Repeats: 0
Total number of images: ~16
Total number of aspect buckets: 1
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

model_id = 'FLUX.1-dev'
adapter_id = 'pytorch_lora_weights.safetensors' # you will have to download this manually
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_id, pipeline.transformer)
wrapper.merge_to()

prompt = "photo of a woman sitting on a bscspc barcelona knoll chair"

pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
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")