Changelog:

V2.0_pynoise: Trained using DAdaptLion optimizer, with multi-resolution noise and minimal SNR gamma enabled. Smaller network_dim/network_alpha (32/16), optimized size of model (36.12M), and reduced overfitting.

V1.0: Initial release. model trained using DAdaptAdam optimizer, without applying multi-resolution noise.

变更日志：

V2.0_pynoise：使用DAdaptLion优化器训练，添加多分辨率噪声，启用最小信噪比伽马值。减小了dim/alpha值（32/16）和文件体积（36.12M），减轻了过拟合现象

V1.0：首发版本，使用DAdaptAdam优化器训练，未添加多分辨率噪声

————————————————————————————————————————

LoRA model of Shizuka Mogami (最上静香) from media-mix project THE IDOLM@STER MILLION LIVE! (アイドルマスターミリオンライブ！).

The trigger word is shizuka mogami (million live). Considering the Second Hair Style (SHS) card and alternative hair color in Mirishita, this model don't fix Shizuka's features of hair and eyes. You can reproduce the Shizuka's style of game card illustrations by adding following tags:

*Original style of Shizuka:

black hair, long hair, blue eyes, blunt bangs, sidelocks, half updo, hime cut, hair ribbon

*Alternative hair style (SHS, ponytail) of Shizuka:

black hair, long hair, blue eyes, blunt bangs, sidelocks, high ponytail

Of course, you can change Shizuka's features of hair and eyes at your own. Try and enjoy it!

(P.S. I think the size of breasts can be easily adjusted by tags e.g. flat chest / tiny tits / small breasts / medium breasts / large breasts / huge breasts / gigantic breasts and thier various weights, so I delete the tag small breasts during captioning, meaning that the small size of breasts has become an inherent feature bound to the character.)

The LoRA weight of samples is 0.8; you can also try other weights. The greater the weight, the closer the character style is to the game card of THE IDOLM@STER MILLION LIVE! THEATER DAYS (ミリシタ) and the new tachie drawn after the release of the game.

Note: I load my LoRA model through the additional network WebUI plugin, so the prompt of sample illustrations don't contain <lora:XXX:X> tag.

I'll appreciate it if you're willing to use this LoRA model and share AI arts to other people. My models and illustrations are free for everyone, and if you like them, please support me at Patreon or 爱发电.

————————————

跨媒体企划《偶像大师 百万现场》的角色最上静香的LoRA模型

模型的触发词为 shizuka mogami (million live)。考虑到游戏的第二发型卡（SHS）和异色发色卡，本模型没有固定静香的头发和眼睛特征，您可以通过添加以下标签来复刻游戏卡面中静香的风格：

*原有卡面的静香:

black hair, long hair, blue eyes, blunt bangs, sidelocks, half updo, hime cut, hair ribbon

*第二发型卡（SHS）的静香:

black hair, long hair, blue eyes, blunt bangs, sidelocks, high ponytail

当然，你也可以自由改变静香的头发和眼睛特征，敬请享受！

示例图的LoRA权重为0.8，您也可以尝试其他权重值。权重越大，角色风格越接近游戏《偶像大师 百万现场 剧场时光》（MLTD）游戏卡面以及该游戏发行后新绘制的立绘

我使用的是WebUI的可选附加网络（additional network）插件，所以示例图的prompt里并没有<lora:XXX:X>的tag，还请注意

如果您愿意使用本LoRA模型并将作品分享给其他人，本人将不胜感激。本人制作的所有模型和绘画皆免费提供，如果觉得还不错的话，欢迎在Patreon或爱发电上支持我，非常感谢。