waveshaker/dataset_collections

---

training dataset for 🧊 waveshaker/loras

Note: The prefixes '1_***' and '5_***' in folder names indicate the repeat number per epoch during training (e.g., 5_*** will be repeated 5 times).

注意：文件夹名称中的前缀 '1_***' 和 '5_***' 代表训练时每个 epoch 的重复次数（例如 5_*** 表示重复 5 次）。

---

Usage Disclaimer

Educational & Research Purpose Only: This repository hosts image-text datasets intended for training character LoRA models across various base models (including Pony, Anima, Flux 1 Dev, Krea2, and SD1.5). All data is provided solely for academic exchange, technical testing, and personal research into image generation algorithms and dataset curation.

Copyright & Data Sourcing: The images and annotations in this repository are collected and pre-processed for experimental training purposes. Users must strictly adhere to local copyright laws and intellectual property regulations. It is strictly forbidden to redistribute or use this dataset for commercial purposes, or to infringe upon the privacy, reputation, or portrait rights of any individuals depicted.

Takedown Notice: This dataset is shared for non-commercial research purposes. If you are a copyright owner or an authorized agent and believe that any content in this repository infringes upon your rights, please contact me directly. I will promptly review and remove the infringing material upon verification, without any delay.

Content Compliance Notice: Please be aware that certain datasets may contain unfiltered or diverse prompts. Users are entirely responsible for auditing, filtering, and ensuring the compliance of any models trained using this data.

Limitation of Liability: As the creator/curator of this experimental dataset, I take no responsibility for any direct or indirect legal liabilities, losses, or disputes arising from the use of this repository. Any models trained using this dataset, and any images subsequently generated by those models, are the sole responsibility of the user.

---

使用免责声明

仅限教育与研究目的：本仓库托管了图文数据集，旨在用于跨多种基础模型（包括 Pony、Anima、Flux 1 Dev、Krea2 和 SD1.5）训练角色 LoRA 模型。所有数据仅供学术交流、技术测试以及个人对图像生成算法和数据集整理的研究使用。

版权与数据来源：本仓库中的图像和标注是为实验性训练目的而收集和预处理的。用户必须严格遵守当地的版权法律和知识产权法规。严禁将本数据集用于商业目的进行再分发或使用，也严禁侵犯所描绘任何个人的隐私、名誉或肖像权。

侵权通知与删除：本数据集仅限非商业研究分享。如果您是版权所有者或授权代理人，并认为本仓库中的任何内容侵犯了您的权利，请直接与我联系。核实后，我将立即审查并毫不拖延地删除相关的侵权内容。

内容合规提示：请注意，某些数据集可能包含未经筛选或多样化的提示词。用户需全权负责对使用此数据训练出的任何模型进行审查、过滤并确保其合规性。

责任限制：作为此实验性数据集的创建者/整理者，我对因使用本仓库而产生的任何直接或间接的法律责任、损失或纠纷概不负责。使用此数据集训练出的任何模型，以及这些模型随后生成的任何图像，均由用户自行承担全部责任。