IMPORTANT: Put "weapon" in negative promptsTrigger word(s): ctiankrulc1fer, light blue hair,Clothing: white capelet, red necktie, underbust, blue skirt, center frills, frilled skirt, long sleeves, black pantyhose,Battlesuit: white unitard, covered navel, collar, white sleeves, black thighhighs, elbow gloves, black gloves,commissions: https://ko-fi.com/ctian 😼ExamplePositive prompt ✅:score_9, score_8_up, score_7_up, source_anime, masterpiece, 1girl, ctiankrulc1fer, light blue hair, blue eyes, white capelet, red necktie, underbust, blue skirt, center frills, frilled skirt, long sleeves, black pantyhose, leaning forward, facing viewer, looking at viewer, outdoors, city, depth of field, sunny, day,Negative prompt ❌:sketch, 3d, pouch, weaponRecommended 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 github4. Train LoRa in kohya ss locally github5. Test epochs using xyz grid in controlnet to ensure LoRa is not overfitted/underfitted.6. Submit model to CivitAI profiledo not ask me training parameters, check lora metadata for training parametersFAQ:Q: why not use character name as trigger word?A: I use unique token to prevent other things from being prompted when using triggerwordQ: do you have other account in other websites?A: yes, tensor, pixaiQ: what external generator do you use?A: a111 mostly, comfyui for img2vid
