Sd35m Sfwbooru Lycoris

🚀 sd35m-sfwbooru-lycoris

這是一個基於 stabilityai/stable-diffusion-3.5-medium 的 LyCORIS 適配器。它能夠在圖像生成任務中，基於基礎模型生成更符合特定需求的圖像，為圖像生成領域帶來更多可能性。

🚀 快速開始

推理示例

以下是使用該適配器進行推理的 Python 代碼示例：

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 = 'stabilityai/stable-diffusion-3.5-medium'
adapter_repo_id = 'bghira/sd35m-sfwbooru-lycoris'
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-realistic image of a cat"
negative_prompt = 'blurry, cropped, ugly'

## 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
model_output = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=30,
    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=3.2,
    skip_guidance_layers=[7, 8, 9],
).images[0]

model_output.save("output.png", format="PNG")

驗證設置

• CFG: 3.2
• CFG 縮放: 0.0
• 步數: 30
• 採樣器: FlowMatchEulerDiscreteScheduler
• 種子: 42
• 分辨率: 1024x1024
• 跳過層引導: skip_guidance_layers=[7, 8, 9]

注意：驗證設置不一定與訓練設置相同。

訓練設置

LyCORIS 配置

{
    "algo": "lokr",
    "multiplier": 1.0,
    "full_matrix": true,
    "linear_alpha": 1,
    "factor": 16,
    "apply_preset": {
        "target_module": [
            "Attention"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 6
            }
        }
    }
}

數據集

sfwbooru-crop

• 重複次數: 0
• 圖像總數: 363920
• 縱橫比桶總數: 1
• 分辨率: 1.048576 兆像素
• 裁剪: 是
• 裁剪樣式: 隨機
• 裁剪縱橫比: 正方形
• 用於正則化數據: 否

指數移動平均 (EMA)

SimpleTuner 會生成 EMA 權重的 safetensors 變體和一個 pt 文件。safetensors 文件用於推理，pt 文件用於繼續微調。EMA 模型可能會提供更全面的結果，但與完整模型相比，由於它是模型權重的衰減平均值，通常會感覺訓練不足。

你可以在以下圖庫中找到一些示例圖像：

文本編碼器未經過訓練，你可以重用基礎模型的文本編碼器進行推理。