exploring a hybrid dataset approach for LoRA training. By Combining manual and automatic tags offers a powerful way to refine control and achieve nuanced outputs. Here's a breakdown of the key points:
Benefits:
* Leverage strengths of both approaches: Automatic tags provide a convenient starting point, while manual tags allow for precise control and artistic vision.
First dataset containing 50 images of trippy and abstract fractal on a dual style and pruned data, for unlimited flexibility trigger words are not needed unless you run into a Lora conflict,
1. 转载模型仅供学习与交流分享，其版权及最终解释权归原作者。
2. 模型原作者如需认领模型，请通过官方渠道联系海艺AI工作人员进行认证。我们致力于保护每一位创作者的权益。 点击去认领
