--- license: other base_model: "FLUX.1-dev" tags: - flux - flux-diffusers - text-to-image - diffusers - simpletuner - not-for-all-audiences - lora - template:sd-lora - lycoris inference: true --- # growwithdaisy/dsycam_20241115_133438 This is a LyCORIS adapter derived from [FLUX.1-dev](https://huggingface.co/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](#training-settings). The text encoder **was not** trained. You may reuse the base model text encoder for inference. ## Training settings - Training epochs: 0 - Training steps: 500 - Learning rate: 0.0001 - Max grad norm: 2.0 - Effective batch size: 16 - Micro-batch size: 2 - 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: ```json { "algo": "lokr", "multiplier": 1, "linear_dim": 1000000, "linear_alpha": 1, "factor": 12, "init_lokr_norm": 0.001, "apply_preset": { "target_module": [ "FluxTransformerBlock", "FluxSingleTransformerBlock" ], "module_algo_map": { "Attention": { "factor": 12 }, "FeedForward": { "factor": 6 } } } } ``` ## Datasets ### mnmlsmo_architecture_photography_style-512 - Repeats: 0 - Total number of images: ~10456 - Total number of aspect buckets: 7 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_architecture_photography_style-768 - Repeats: 0 - Total number of images: ~9000 - Total number of aspect buckets: 10 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_architecture_photography_style-1024 - Repeats: 2 - Total number of images: ~3960 - Total number of aspect buckets: 1 - Resolution: 1.048576 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_art_photography_style-512 - Repeats: 0 - Total number of images: ~448 - Total number of aspect buckets: 6 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_art_photography_style-768 - Repeats: 0 - Total number of images: ~416 - Total number of aspect buckets: 7 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_art_photography_style-1024 - Repeats: 1 - Total number of images: ~232 - Total number of aspect buckets: 9 - Resolution: 1.048576 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_furniture_photography_style-512 - Repeats: 0 - Total number of images: ~3888 - Total number of aspect buckets: 13 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_furniture_photography_style-768 - Repeats: 0 - Total number of images: ~3352 - Total number of aspect buckets: 13 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_furniture_photography_style-1024 - Repeats: 1 - Total number of images: ~1728 - Total number of aspect buckets: 4 - Resolution: 1.048576 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_homewares_photography_style-512 - Repeats: 0 - Total number of images: ~1096 - Total number of aspect buckets: 6 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_homewares_photography_style-768 - Repeats: 0 - Total number of images: ~1072 - Total number of aspect buckets: 3 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_homewares_photography_style-1024 - Repeats: 1 - Total number of images: ~520 - Total number of aspect buckets: 2 - Resolution: 1.048576 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_interiors_photography_style-512 - Repeats: 0 - Total number of images: ~1336 - Total number of aspect buckets: 4 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_interiors_photography_style-768 - Repeats: 0 - Total number of images: ~1312 - Total number of aspect buckets: 5 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_interiors_photography_style-1024 - Repeats: 1 - Total number of images: ~800 - Total number of aspect buckets: 1 - Resolution: 1.048576 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_lighting_photography_style-512 - Repeats: 0 - Total number of images: ~504 - Total number of aspect buckets: 5 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_lighting_photography_style-768 - Repeats: 0 - Total number of images: ~504 - Total number of aspect buckets: 5 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_lighting_photography_style-1024 - Repeats: 0 - Total number of images: ~320 - Total number of aspect buckets: 4 - Resolution: 1.048576 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_moods_photography_style-512 - Repeats: 0 - Total number of images: ~680 - Total number of aspect buckets: 3 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_moods_photography_style-768 - Repeats: 0 - Total number of images: ~680 - Total number of aspect buckets: 3 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_moods_photography_style-1024 - Repeats: 1 - Total number of images: ~352 - Total number of aspect buckets: 2 - Resolution: 1.048576 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_technology_photography_style-512 - Repeats: 0 - Total number of images: ~680 - Total number of aspect buckets: 4 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_technology_photography_style-768 - Repeats: 0 - Total number of images: ~680 - Total number of aspect buckets: 4 - Resolution: 0.589824 megapixels - Cropped: False - Crop style: None - Crop aspect: None - Used for regularisation data: No ### mnmlsmo_technology_photography_style-1024 - Repeats: 1 - Total number of images: ~376 - Total number of aspect buckets: 3 - Resolution: 1.048576 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 = 'FLUX.1-dev' adapter_repo_id = 'playerzer0x/growwithdaisy/dsycam_20241115_133438' 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") ```