Sd35m Photo 1mp Prodigy

🚀 sd35m-photo-1mp-Prodigy

sd35m-photo-1mp-Prodigy是一个基于stabilityai/stable-diffusion-3.5-medium的LyCORIS适配器。它可用于图像生成任务，在文本到图像、图像到图像等转换场景中发挥作用。

🚀 快速开始

推理使用示例

以下是使用该适配器进行推理的Python代码示例：

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-photo-1mp-Prodigy'
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 = 'ugly, cropped, blurry, low-quality, mediocre average'

## Optional: quantise the model to save on vram.
## Note: The model was quantised during training, and so it is recommended to do the same 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=16,
    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")

✨ 主要特性

• 基于stabilityai/stable-diffusion-3.5-medium模型，可用于文本到图像、图像到图像等转换。
• 文本编码器未进行训练，推理时可复用基础模型的文本编码器。
• 提供了验证设置和训练设置的详细信息。
• 生成了EMA权重的safetensors变体和pt文件，可分别用于推理和继续微调。

🔧 技术细节

验证设置

• CFG: 3.2
• CFG Rescale: 0.0
• Steps: 16
• Sampler: FlowMatchEulerDiscreteScheduler
• Seed: 42
• Resolution: 1024x1024
• Skip-layer guidance: skip_guidance_layers=[7, 8, 9]

需注意，验证设置不一定与训练设置相同。

训练设置

• 训练轮数: 114
• 训练步数: 230
• 学习率: 5e-05
  • 学习率调度: 恒定
  • 热身步数: 500
• 最大梯度值: 2.0
• 有效批次大小: 3
  • 微批次大小: 3
  • 梯度累积步数: 1
  • GPU数量: 1
• 梯度检查点: 启用
• 预测类型: flow_matching (额外参数=['shift=3.0'])
• 优化器: optimi-lion
• 可训练参数精度: Pure BF16
• 基础模型精度: int8-quanto
• 字幕丢弃概率: 0.0%

LyCORIS配置

{
    "bypass_mode": true,
    "algo": "lokr",
    "multiplier": 1.0,
    "full_matrix": true,
    "linear_dim": 10000,
    "linear_alpha": 1,
    "factor": 4,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "FeedForward": {
                "factor": 4
            },
            "Attention": {
                "factor": 2
            }
        }
    }
}

数据集

cheechandchong

• 重复次数: 0
• 图像总数: 4
• 纵横比桶总数: 1
• 分辨率: 1024 px
• 裁剪: 是
• 裁剪样式: 随机
• 裁剪纵横比: 方形
• 用于正则化数据: 否

指数移动平均 (EMA)

SimpleTuner生成了EMA权重的safetensors变体和pt文件。safetensors文件用于推理，pt文件用于继续微调。EMA模型可能会提供更全面的结果，但与完整模型相比，通常会感觉训练不足，因为它是模型权重的衰减运行平均值。

📄 许可证

许可证类型：other