ハイレグの角度を強化する、スーパーハイレグのAnima用Loraです。This is a Lora for Anima that enhances the angle of the high-leg design, achieving a super high-leg look.v1.5：preview2をベースにし、自然言語キャプションを追加して学習v1.4：2/84を生成画像で学習。効果自体は減ったかも知れませんが、生成画像特有のノイズとバージョンアップで積み重なった過剰な皺が解消…したはず。v0.1： 全てillustrious系の生成画像で学習。illustrious版のv3.6と同一のデータセット。<hr /><h2>設定</h2>トリガーワード： <code>super highleg</code>推奨ウェイト： 1学習対象の衣装：<ul><li>leotard</li><li>one-piece swimsuit</li></ul>上記以外の衣装に適用したい場合には<code>revealing clothes</code>, <code>leotard</code>, <code>swimsuit</code>などのタグと併用することで、ある程度元の意匠を保ちながらもハイレグ衣装となります。効果を強化する場合は<code>(super highleg:2)</code> のようにタグのウェイトを上げるか、LoRAのウェイトを上げてください。<hr /><h2 id="settings">Settings</h2>Trigger Word: <code>super highleg</code>Recommended Weight: 1Target Outfits for Learning:<ul><li>leotard</li><li>one-piece swimsuit</li></ul>To apply this to outfits other than those listed above, combine it with tags like <code>revealing clothes</code> , <code>leotard</code>, or <code>swimsuit</code> This will create a high-cut outfit while preserving some of the original design.To enhance the effect, increase the tag weight <code>(super highleg:2)</code> or raise the LoRA weight.<hr /><h2>概念イメージ</h2>作者の脳内イメージ（普通のハイレグでは角度が足りない）黄色線ぐらいのラインになってほしいという願望The author's mental image (a regular high-leg cut doesn't have enough angle)The desire for it to be a line like the yellow one<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0ff86ab2-a125-46b5-bf30-daed354e73b1/original=true/0ff86ab2-a125-46b5-bf30-daed354e73b1.jpeg" /><hr />The CircleStone Model is licensed by CircleStone Labs LLC under the CircleStone Non-Commercial License. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL. <a target="_blank" rel="ugc" href="https://huggingface.co/circlestone-labs/Anima/resolve/main/LICENSE.md">circlestone-labs-non-commercial-license</a>

Anima LoRA super highleg

images: 82resolution = "1024,1024"optimizer_type = "optimizer.emosens.EmoSens"learning_rate = 1lr_scheduler = "constant"train_batch_size = 4network_dim = 8network_alpha = 8

Files with SHA256: dc414e956224a2eb0723af439426a61c921b951701529493d37aa7f194cc5230