<h3 id="important:-read-below-7x1uosjre"><a target="_blank" rel="ugc" href="https://ko-fi.com/ctian"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/cf614846-63c8-4596-a5fd-f5962d69ae75/width=525/cf614846-63c8-4596-a5fd-f5962d69ae75.jpeg" /></a>IMPORTANT: READ BELOW</h3>Trigger word(s): ctianhimesaka, Clothing: sailor dress, neck bow, white shirt, sailor collar, long sleeves, Example Positive prompt:<pre><code>score_9, score_8_up, score_7_up, source_anime, masterpiece, 1girl, ctianhimesaka, sailor dress, neck bow, white shirt, sailor collar, long sleeves, looking at viewer,</code></pre>Negative prompt:<pre><code>sketch, 3d</code></pre>Recommended resolution, aspect ratio for sdxl, pony768 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, procedure1. 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>

Himesaka Noa 姫坂 乃愛  PONY - Watashi ni Tenshi ga Maiorita! Wataten!: An Angel Flew Down to Me

<ul><li>20 extra images added to training dataset</li><li>retrained on base pony model</li><li>improved captioning on training dataset</li></ul>

HimesakaNoaPONYXLversion2.safetensors

v2.0

v1.0

<h3 id="himesaka-noa-from-watashi-ni-tenshi-ga-maiorita!-wataten!pony-lora-gwks1xw8j">Himesaka Noa 姫坂 乃愛 from Watashi ni Tenshi ga Maiorita! Wataten! PONY Lora</h3>Trigger words: ctianhimesaka Clothing: sailor dress, neck bow, white shirt, sailor collar, long sleeves,Consider buying me a kofi ? <a target="_blank" rel="ugc" href="https://ko-fi.com/ctian">https://ko-fi.com/ctian</a>768 x 1344: Vertical (9:16)915 x 1144: Portrait (4:5)1024 x 1024: square 1:11182 x 886: Photo (4:3)1254 x 836: Landscape (3:2)1365 x 768: Widescreen (16:9)1564 x 670: Cinematic (21:9)

&lt;ul&gt;&lt;li&gt;&lt;p&gt;20 extra images added to training dataset&lt;/p&gt;&lt;/li&gt;&lt;li&gt;&lt;p&gt;retrained on base pony model&lt;/p&gt;&lt;/li&gt;&lt;li&gt;&lt;p&gt;improved captioning on training dataset&lt;/p&gt;&lt;/li&gt;&lt;/ul&gt;

Files with SHA256: bc13ecba84f6eeadb5bd2e7a3d86a0ba726970c72d8c697d6c7d0081fa05b6b5