Version V2:

The V2 version has changed the basic model for training, and according to the comments, the anime model of sd1.5 has been used for training. At the same time, better background images were used, totaling 1100 images. The model effect of the V2 version can be presented normally on different checkpoint models, and checkpoint models such as anything, cetus, OrangeMix, ghoatMix, etc. have been tested.

This version has improved image logic, color effects, and stylization, and also has good effects on animated graphics. It is recommended to use it.

Issues to be improved:

Due to the excessive number of training images, the processing of prompt words is relatively sloppy and not handled very meticulously.

2. For wires with the characteristic expression of the Story series, the learning effect of shadow processing is weakened.

版本更新V2：

V2版本更换了训练的基本模型，根据评论的建议，采用sd1.5的anime模型进行训练。同时采用了更好的背景图片，共计1100张图片。V2版本的模型效果可以在不同checkpoint模型上正常呈现，已经测试anything,cetus,OrangeMix,ghoatMix等checkpoint模型。

该版本在图像逻辑，颜色效果，风格化上均有提升，在图生图上也有良好的效果，更建议使用。

待改进的问题：

1. 由于训练图片数量过多，提示词的处理相对马虎，没有非常细致的处理。

2. 对于有物语系列特色表达的电线，阴影处理效果上学习效果减弱。

3. 对人物的表现能力较差，更加强调环境的表达

Introduction（中英文简介）:

This is my first stylized Lora that I've created. The theme is inspired by the Monogatari Series anime style, focusing on landscapes such as streets and cities, and it does not include any characters. I created this Lora because of my love for the Monogatari Series. Civitai was lacking in this particular style of artwork. The Monogatari Series has a distinct and unique style of its own, and I would have felt regretful if there wasn't a Lora in this style. This Lora mainly uses screenshots from the anime adaptations of "Bakemonogatari," "Nisemonogatari," and "Monogatari Series: Second Season." However, it's not yet complete, and I plan to add artworks from other installments of the Monogatari Series in the future. This Lora does not include "Owarimonogatari," as I plan to train a separate Lora for that specific installment.

这是我炼制的第一个风格化lora.主题为 物语系列动画风格，不包含人物，以街道，城市等风景的风格为主。炼制这个Lora的原因是出于对《物语系列》的热爱，而civitai缺少该风格的作品，物语系列风格鲜明有自己的特色，如果没有这种风格的lora我会感到非常遗憾。该Lora的主要使用了《化物语》，《倾物语》，《恋物语》的相关动画截图，还不够完善，后续会继续添加其他物语系列的作品。该lora不包含《伤物语》，《伤物语》的lora我会另外训练。

Suggestions for Use:

base model:

This Lora is trained on the base model YabukiKentaroStyle. If using this Lora, it is recommended to use it in conjunction with YabukiKentaroStyle for optimal results. While other models also produce Lora effects, their performance is relatively inferior.

optimization：

it is recommended to use the DMP++2M SDE optimizer.

Parameters:

0.6-1：The primary alterations focus on lighting and color style, creating a sense of lines

1-1.2：The composition leans towards mid to distant perspectives, featuring strong contrasting colors and enhanced linearity.

above 1.5: Not recommended to use versions above 1.5. While the style becomes more pronounced, the visual coherence of the image deteriorates.

Note:

Terminology such as "utility poles," "traffic lights," and "red street lamps" will have a significant impact on the composition. This is related to the materials I have chosen. This issue will be addressed in future versions.

基本模型：

本lora是在基本模型yabukiKentaroStyle上训练的，如果使用该lora建议与yabukiKentaroStyle配合使用，其他模型也有lora效果但效果相对较差。

优化器：DMP++2M SDE

参数：

0.6-1：主要改变光照和颜色风格，具有线条感。

1.0-1.2：画面相对倾向于中远景，具有强对比色，线条感更加明显

不建议使用1.5及其以上，会导致画面逻辑崩坏，但是风格更鲜明

注意：

电线杆，信号灯，红路灯等词汇会对画面产生较大影响（这与我选用的材料相关，后续版本会解决该问题）