<h3 id="junko-takei-pony-strike-witches-bgw7c3h24">Junko Takei 竹井醇子 PONY - Strike Witches ストライクウィッチーズ</h3>Full 11 Character collection <a target="_blank" rel="ugc" href="https://civarchive.com/collections/2226666">Strike Witches</a>kofi commission <a target="_blank" rel="ugc" href="https://civarchive.com/user/Voyager_a">@Voyager_a</a> Trigger words: ctianjunko, Clothing: white uniform, swimsuit under clothes, Animal ears: dog ears, dog tail,commission: <a target="_blank" rel="ugc" href="https://ko-fi.com/ctian">https://ko-fi.com/ctian</a> 😼 Example Positive prompt ✅:<pre><code>score_9, score_8_up, score_7_up, source_anime, masterpiece, 1girl, ctianjunko, white uniform, swimsuit under clothes, dock, outdoors, ocean, depth of field, looking at viewer, facing viewer, dog ears, dog tail, </code></pre>Negative prompt ❌:<pre><code>sketch, 3d</code></pre>Recommended resolution, aspect ratio for sdxl, pony 📷768 x 1344: Vertical (9:16)832 x 1216: Portrait (4:5) **1024 x 1024: Square (1:1)1182 x 886: Photo (4:3)1254 x 836: Landscape (3:2)1365 x 768: Widescreen (16:9)Work flow, procedure 🔧1. Collect dataset (booru, anime screencaps, search)2. Clean up dataset using upscalers, inpainting and photoshop3. Tag dataset using booru <a target="_blank" rel="ugc" href="https://github.com/starik222/BooruDatasetTagManager">github</a>4. Train LoRa in kohya ss locally <a target="_blank" rel="ugc" href="https://github.com/bmaltais/kohya_ss">github</a>5. Test epochs using xyz grid in controlnet to ensure LoRa is not overfitted/underfitted.6. Submit model to CivitAI <a target="_blank" rel="ugc" href="https://civarchive.com/user/ctian">profile</a>do not ask me training parameters, check lora metadata for training parameters

Junko Takei 竹井醇子 PONY - Strike Witches ストライクウィッチーズ

<a target="_blank" rel="ugc" href="https://civitai.com/user/Voyager_a">@Voyager_a</a>

Files with SHA256: d6009e47c2abba8fa0f2ccf261623c6d05160d4c0625f60e825c4526bc337004