🚀 Bold-Shadows-Flux-LoRA
Bold-Shadows-Flux-LoRA 是一款圖像生成模型,可基於特定提示詞生成具有鮮明陰影效果的黑白圖像,目前仍在訓練階段。
🚀 快速開始
模型加載
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/Bold-Shadows-Flux-LoRA"
trigger_word = "Bold Shadows"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
觸發詞使用
你應該使用 Bold Shadows
來觸發圖像生成。
模型下載
此模型的權重以 Safetensors 格式提供。
點擊下載,可在 Files & versions 標籤中獲取。
✨ 主要特性
- 能夠生成具有鮮明陰影效果的黑白圖像。
- 基於特定的圖像生成參數進行訓練。
📦 安裝指南
暫未提供具體的安裝步驟,可參考上述快速開始部分的代碼示例進行模型加載。
💻 使用示例
基礎用法
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/Bold-Shadows-Flux-LoRA"
trigger_word = "Bold Shadows"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
prompt = "Bold Shadows, Captured in a black and white monochromatic shot, a woman stands in front of a backdrop of leaves and shadows. She is wearing a black bra, her hair cascades over her shoulders. Her eyes are squinted and her lips are pursed. Her hair is long and cascades in a ponytail, adding a touch of movement to her face. The wall behind her is made up of white tiles, creating a stark contrast to the womans body."
image = pipe(prompt).images[0]
image.save("generated_image.png")
📚 詳細文檔
模型描述
prithivMLmods/Bold-Shadows-Flux-LoRA
圖像參數處理
參數 |
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 & 2440 |
Epoch |
10 |
Save Every N Epochs |
1 |
標註:florence2-en(自然語言 & 英語)
用於訓練的總圖像數:14
最佳尺寸
注意事項
⚠️ 重要提示
該模型仍處於訓練階段。這不是最終版本,可能包含偽影,並且在某些情況下性能不佳。
模型託管
- 託管地址🧨: https://huggingface.co/spaces/prithivMLmods/FLUX-LoRA-DLC
📄 許可證
本模型採用 creativeml-openrail-m 許可證。