<h2>Li Wen (WILL: A Wonderful World) 李雯 (WILL：美好世界)</h2><a target="_blank" rel="ugc" href="https://civarchive.com/models/38103/experimentalalicia-will-a-wonderful-world-will">Alicia (WILL: A Wonderful World) 阿莉西亚 (WILL：美好世界)</a><a target="_blank" rel="ugc" href="https://civarchive.com/models/38957/experimentalpi-will-a-wonderful-world-pi-will">Pi (WILL: A Wonderful World) Pi (WILL：美好世界)</a>Triggers with <code>liwen_will</code>, weight has to be 0.6.This LoRA is an experiment to answer whether I can make a decent LoRA with only 10 copies of the same image, considering that Li Wen unfortunately doesn't have that many artworks to put into the dataset, even with Lofter and TapTap's gallery.The dataset was this:<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/657d4c4d-824a-4428-4b93-0f8f131e5200/width=525/657d4c4d-824a-4428-4b93-0f8f131e5200.jpeg" />As you can see... it's the same image. I didn't used regularization images, but I might as I try to use the AI outputs as inputs for a future v2.Used Cetus-mix and Anypastel for example images.<h2>Usage</h2>General tags: <code>liwen_will</code> works fine already, but you can use <code>pink hair, long hair, orange eyes, long hair</code> if necessary.For her uniform, just use <code>default_outfit, school uniform, black vest, black skirt, yellow ribbon, long sleeves, white sleeves.</code><h2>Known Issues</h2><ul><li>Don't bother using it at a weight higher than 0.6, it'll overfit.</li><li>Sometimes her sleeves will appear as black. Sometimes I tag <code>long sleeves, white sleeves</code> to make it happen less.</li></ul><h2>Training Parameters</h2><ul><li>"ss_text_encoder_lr":"1e-05",</li><li>"ss_total_batch_size":"2",</li><li>"ss_num_epochs":"10",</li><li>"ss_unet_lr":"0.0001",</li><li>"ss_learning_rate":"0.0001",</li><li>"ss_lr_scheduler":"cosine_with_restarts",</li><li>"ss_lr_warmup_steps":"0",</li><li>"ss_max_token_length":"225",</li><li>"ss_mixed_precision":"fp16",</li><li>"ss_network_alpha":"64.0",</li><li>"ss_network_dim":"128",</li><li>{\"n_repeats\": 20, \"img_count\": 10}</li><li>"ss_resolution":"(768, 768)"</li><li>"ss_sd_model_name":"Animefull-final-pruned.ckpt",</li><li>"ss_seed":"42",</li></ul>

Files with SHA256: 1d8664a4109dcb8a49e31f3709ba898e0c5fed86c2198e569a404e5eb9a08a79