🚀 Yellow-Pop-Flux-Dev-LoRA
Yellow-Pop-Flux-Dev-LoRA 是一个文本到图像的模型,基于 LoRA 技术,可生成具有特定风格的图像。该模型目前仍处于训练阶段,能够根据特定的提示词生成相应的图像。
🚀 快速开始
环境设置
import torch
from pipelines import DiffusionPipeline
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
lora_repo = "prithivMLmods/Yellow-Pop-Flux-Dev-LoRA"
trigger_word = "Yellow Pop"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
触发图像生成
你应该使用 Yellow Pop
来触发图像生成。
模型下载
该模型的权重以 Safetensors 格式提供。
点击下载,可在 Files & versions 选项卡中获取。
✨ 主要特性
- 文本到图像转换:根据输入的文本提示生成相应的图像。
- 特定风格:生成具有 “Yellow Pop” 风格的图像。
- LoRA 技术:基于 LoRA 技术进行训练,提高模型效率。
📦 安装指南
文档中未提及具体的安装步骤,可参考上述快速开始部分的代码示例进行环境设置。
💻 使用示例
基础用法
以下是一个简单的使用示例,展示了如何加载模型并进行图像生成:
import torch
from pipelines import DiffusionPipeline
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
lora_repo = "prithivMLmods/Yellow-Pop-Flux-Dev-LoRA"
trigger_word = "Yellow Pop"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
高级用法
文档中未提及高级用法的代码示例,可根据具体需求进一步探索模型的功能。
📚 详细文档
模型描述
prithivMLmods/Yellow-Pop-Flux-Dev-LoRA
属性 |
详情 |
基础模型 |
black-forest-labs/FLUX.1-dev |
实例提示词 |
Yellow Pop |
许可证 |
creativeml-openrail-m |
图像处理参数
参数 |
值 |
参数 |
值 |
LR Scheduler |
constant |
Noise Offset |
0.03 |
Optimizer |
AdamW |
Multires Noise Discount |
0.1 |
Network Dim |
64 |
Multires Noise Iterations |
10 |
Network Alpha |
32 |
Repeat & Steps |
22 & 2600 |
Epoch |
15 |
Save Every N Epochs |
1 |
标注
使用 florence2 - en(自然语言 & 英语)进行标注。
训练使用的总图像数
22 张
最佳尺寸
数据来源
🔧 技术细节
该模型基于 LoRA 技术,结合了 black-forest-labs/FLUX.1-dev
基础模型进行训练。在训练过程中,使用了特定的图像处理参数,如 LR Scheduler、Optimizer 等,以优化模型性能。同时,使用了 22 张图像进行训练,并采用了 florence2 - en 进行标注。
📄 许可证
该模型使用 creativeml-openrail-m
许可证。
⚠️ 重要提示
该模型仍处于训练阶段,这不是最终版本,可能会包含伪影,并且在某些情况下表现不佳。
💡 使用建议
建议在使用时使用 Yellow Pop
作为触发词,以获得最佳的图像生成效果。同时,可根据具体需求调整图像处理参数,以满足不同的应用场景。
模型展示
示例图像
文本提示 |
输出图像 |
'Yellow Pop, A medium-sized bulldog stands in front of a vibrant pink background. The bulldogs face is prominently displayed, with its mouth slightly open. Its eyes are wide open, and its ears are perked up. Its mouth is slightly open, revealing a black nose. Its muzzle is adorned with small black dots, adding a pop of color to the scene. The dogs coat is a light brown, with a purple collared shirt underneath it. The coat has a black collar with a few buttons on it.' |
images/YP1.png |
'Yellow Pop, An eye-level painting of a tree with a yellow base and a thick trunk. The trees leaves are a vibrant shade of green, while the base is a lighter shade of yellow. The background is a deep pink, and the trees shadow is cast on the ground below the tree, creating a striking contrast of colors. The leaves are arranged in a way that creates a sense of depth and dimension, adding a pop of color to the scene.' |
images/YP2.png |
'Yellow Pop, A digital art painting of a woman in a pink jacket and a yellow turtleneck. The womans hair is styled in a bun and she is wearing a pair of pink tinted glasses. Her eyes are blue and her lips are pink. Her hair is pulled back in a ponytail. She is wearing large silver hoop earrings on both sides of her ears. The background of the painting is a vibrant yellow.' |
images/YP3.png |