🚀 Canopus-Clothing-Adp-LoRA
Canopus-Clothing-Adp-LoRA 是一个专注于服装相关图像生成的模型,借助 LoRA 技术,在基于 StableDiffusionXL 基础模型上进行训练,能根据特定服装提示词生成高质量的图像。
🚀 快速开始
环境搭建
pipe = StableDiffusionXLPipeline.from_pretrained(
"-------------xxxxxxxxx----------",
torch_dtype=torch.float16,
use_safetensors=True,
)
(or)
-----------------------------------------------------------
pipe = StableDiffusionXLPipeline.from_pretrained(
"stabilityai/stable-diffusion-xl-base-1.0",
torch_dtype=torch.float16,
use_safetensors=True,
)
pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
pipe.load_lora_weights("prithivMLmods/Canopus-Clothing-Adp-LoRA", weight_name="Canopus-Dress-Clothing-LoRA.safetensors", adapter_name="clth")
pipe.set_adapters("clth")
pipe.to("cuda")
触发提示词
- Front view, capture a urban style, Hoodie, technical materials, fabric small point label on text 'Graytheory', the design is minimal, with a raised collar, fabric is a dark grey, low angle to capture the Hoodie's form and detailing, f/5.6 to focus on the hoodie's craftsmanship, solid grey background, studio light setting
- Front view, capture a urban style, Hoodie, technical materials, fabric small point label on text 'yellow theory', the design is minimal, with a raised collar, fabric is a light yellow, low angle to capture the Hoodie's form and detailing, f/5.6 to focus on the hoodie's craftsmanship, solid grey background, studio light setting
- Front view, capture a urban style, Hoodie, technical materials, fabric small point label on text 'red theory', the design is minimal, with a raised collar, fabric is a light red, low angle to capture the Hoodie's form and detailing, f/5.6 to focus on the hoodie's craftsmanship, solid grey background, studio light setting
- Front view, capture a urban style, Batman Hoodie, technical materials, fabric small point label on text 'Black theory', the design is minimal, with a raised collar, fabric is a Light red, low angle to capture the Hoodie's form and detailing, f/5.6 to focus on the hoodie's craftsmanship, solid grey background, studio light setting, with batman logo in the chest region of the t-shirt
- Front view, capture a urban style, Superman Hoodie, technical materials, fabric small point label on text 'Blue theory', the design is minimal, with a raised collar, fabric is a Light yellow, low angle to capture the Hoodie's form and detailing, f/5.6 to focus on the hoodie's craftsmanship, solid grey background, studio light setting, with batman logo in the chest region of the t-shirt
- Front view, capture a urban style, Gta Hoodie, technical materials, fabric small point label on text 'Green theory', the design is minimal, with a raised collar, fabric is a Rose, low angle to capture the Hoodie's form and detailing, f/5.6 to focus on the hoodie's craftsmanship, solid grey background, studio light setting, with batman logo in the chest region of the t-shirt
触发参数
参数 |
详情 |
提示词 |
Front view, capture a urban style, Superman Hoodie, technical materials, fabric small point label on text 'Blue theory', the design is minimal, with a raised collar, fabric is a Light yellow, low angle to capture the Hoodie's form and detailing, f/5.6 to focus on the hoodie's craftsmanship, solid grey background, studio light setting, with batman logo in the chest region of the t-shirt |
采样器 |
euler |
触发关键词
你应该使用 Dress, Cloth, Cloth style, Material Style
来触发图像生成。
✨ 主要特性
- 专注服装生成:该模型更适合基于服装的上下文提示,能够生成各种风格和材质的服装图像。
- 持续优化:模型仍处于训练阶段,后续会不断改进以提升性能和减少瑕疵。
📦 安装指南
模型权重以 Safetensors 格式提供。
点击下载,可在“Files & versions” 标签中获取。
📚 详细文档
模型描述
prithivMLmods/Canopus-Clothing-Adp-LoRA
图像生成参数如下:
属性 |
详情 |
LR调度器 |
constant |
噪声偏移 |
0.03 |
优化器 |
AdamW |
多分辨率噪声折扣 |
0.1 |
网络维度 |
64 |
多分辨率噪声迭代次数 |
10 |
网络Alpha |
32 |
重复次数与步数 |
20 & 1000 |
训练轮数 |
10 |
每N轮保存一次 |
1 |
注意事项
⚠️ 重要提示
该模型仍处于训练阶段,这并非最终版本,可能存在瑕疵,在某些情况下表现不佳。
📄 许可证
本模型使用的许可证为 creativeml-openrail-m。