<h3>YueZheng Ling (乐正绫) LoRA</h3>An anime LoRA model for Chinese vocaloid singer YueZheng Ling. Tested on <a rel="ugc" href="https://civarchive.com/models/4468/counterfeit-v25">Counterfeit-V2.5</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/7189/school-anime">schoolanime</a>, <a rel="ugc" href="https://civarchive.com/models/11866/meinapastel">meinapastel_V2</a>, should works on other anime models too. If you have any question, please leave a comment. Welcome to share your pictures!乐正绫的LoRA模型, 在<a target="_blank" rel="ugc" href="https://civarchive.com/models/4468/counterfeit-v25">Counterfeit-V2.5</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/7189/school-anime">schoolanime</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/11866/meinapastel">meinapastel_V2</a>上已经测试过, 其他模型应该也能用. 如果生成了好看的图一定要发出来 我也要看!!! Example images are generated with <a target="_blank" rel="ugc" href="https://civarchive.com/models/7189/school-anime">schoolanime</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/11866/meinapastel">meinapastel_V2.</a>示例图使用<a target="_blank" rel="ugc" href="https://civarchive.com/models/7189/school-anime">schoolanime</a>和<a target="_blank" rel="ugc" href="https://civarchive.com/models/11866/meinapastel">meinapastel_V2</a>生成<h3> Use </h3>prompt: yuezhengling, &lt;lora:yuezhengling:0.6&gt; recommended lora weight: 0.6-0.8推荐lora权重: 0.6-0.8 <h3>Train</h3>Base model: NovelAI dataset: 25 images epoch: 10 (each epoch 1000 steps)LION Optimizer 基于NovelAI训练,数据集使用25张图片,10个epoch, 每个epoch跑了1000step, 使用LION Optimizer, 这次认真处理了下数据集，有空会按照同样的方法更新以前发布的<s>粗制滥造的</s>模型

Files with SHA256: 85dd9c4fd939bb07317751279cb230186db155b0799fa873af4cf023cdb480bb