1.0版本

很抱歉我的英语不是很好。

I'm sorry my English is not very good .

使用此lora可以使画面层次更深，推荐权重不低于1.2使用，与人物lora和色彩饱和度较高的模型搭配使用更加，如果你不喜欢图片中出现文字，请在negtive prompt中加入“text”并调高权重。1.0版本的触发词暂定为“sqmr/(chahua/)”。由于目前训练集较少，画风不是特别明显。已经在加急训练新版本了。

Using this LORA can make the image more layered, with a recommended weight of no less than 1.2. It is more suitable for use with character LORA and models with higher color saturation.If you don't like the text appearing in the image, please add "text" in the negative prompt and increase the weight.The trigger word for version 1.0 is tentatively set to "sqmr/(chahua/)". Due to the small number of training sets at present, the style of painting is not particularly obvious. The new version is already being trained urgently.

这是我第一次训练模型，有所欠缺的地方欢迎大家指出。如果你喜欢我的模型，请给我五星好评，如果你不喜欢，也请不要打一星。

This is my first time to train the model, and you are welcome to point out some deficiencies. If you like my model, please give me five-star praise, if you don't like it, please don't give me bad reviews.

1.1版本

嗨，大家好。本次更新的1.1版本相比与1.0版本增加了作者大量早期的作画作为训练集，并且对参数和细节进行了很多调整，画风更加全面也更明显了。

Hi, everybody. Compared with version 1.0, the 1.1 version of this update adds a large number of early paintings as a training set, and makes a lot of adjustments to parameters and details, making the painting style more comprehensive and obvious.

下面是使用说明：模型权重推荐1左右，根据你想要的效果具体调整。触发词为sqmr1。

由于深崎暮人老师的作品时间跨度较大，风格变化也很大，如果你想要生成一张他早期风格的作画，这边推荐不要使用色彩饱和度高和风格特征强烈的模型，不要使用类似“best quality”的关键词。

Here are the instructions: the weight of the model is recommended about 1, which can be adjusted according to the effect you want. The trigger word is "sqmr1".
Because of the wide time span and great changes in style of Mr. Miyazaki's works, if you want to generate a painting of his early style, it is recommended that you do not use models with high color saturation and strong style features, and do not use keywords like "best quality".

下面分享一下我常用的参数。step：30，sampler：DMP++ 2M Karras，Width*Height：512*768。

Let's share some of my commonly used parameters. Step:30, sampler:DMP++ 2M Karras, Width*Height:512*768.

目前版本的问题：由于加入了很多早期的作品，1.1版本相比于1.0版本在画面色彩表现上略逊一筹。还有就是1.1版本的手部描绘比较差容易扭曲，建议配合高清修复一起使用，如果你想要更加艳丽的色彩和更准确的手部描绘，这里推荐使用1.0版本。但是就我个人而言更推荐1.1版本。

The problem with the current version: due to the addition of many earlier works, version 1.1 is slightly inferior to version 1.0 in terms of color performance. In addition, version 1.1 of the hand description is relatively poor and easy to distort，recommended to use it with Hires.fix, . If you want more colorful colors and more accurate hand description, version 1.0 is recommended here.Overall, I prefer version 1.1.But personally, I prefer version 1.1.

这次更新爆肝了好久，如果你喜欢这个模型的话请务必给我五星好评，蟹蟹辣。后续应该还有一或者两次更新。

This update takes a long time, if you like this model, please do give me five-star praise, thx. There should be one or two updates later.

2.0版本

大家好，好久不见了。这次带来的是画风lora的2.0更新，我在这次的新版本中使用了很多我新学到的技巧，使它相较于1.0和1.1版本有很大的提升。

使用说明：

weight：0.7-0.9,过高的权重容易带来手部崩坏，但是不同模型的效果不尽相同

改动说明：

更换了训练的底模，使模型有更好的泛用性

调整了训练集，从2000张图片中挑选了画风较为明显的120张图片。根据1.1版本的反馈，大家并不喜欢太久远的画风，所以这次没有使用很多古早时期的图片作为训练集。

为训练集增加了tag，我在网上查阅了很多资料，对于画风训练有两种看法。一种是只保留触发词，另一种是为能被定义的全部元素都打上标签，这次我尝试了后者，效果差强人意，还有优化空间（其实是太烂了不想手打tag）。

优化说明：

相比于1.1版本，画风更为明显，即使只使用1girl也能有很明显的效果

修复了1.1版本色彩不够鲜艳，画面色调较暗的问题，而且进行了增强，2.0版本的画面有更强的色彩对比度。

支持512*512、512*768、768*512三种分辨率。

缺点：

手的部分容易崩坏，尝试了很多种方法都没有较好的效果。可以尝试多增加一点负面词降低手部崩坏的概率

这个模型我应该会继续优化到3.0，如果觉得我做的还不错的话请点一下❤，：D

以下是机翻，如果有翻译错误敬请见谅/The following is translated by machine. Please forgive me if there are any translation errors.

Hello everyone, long time no see. This time I bring you the 2.0 update of the Lora style, and I have used many new techniques that I learned in this new version, which significantly improves upon versions 1.0 and 1.1.

Instructions for use:
Weight:0.7-0.9. excessive weight is easy to cause hand collapse, but the effect of different models is different.

Changes:

Replaced the base model used for training to improve its general usability.

Adjusted the training set by selecting 120 images with more distinct artistic styles from a pool of 2000 images. Based on feedback from version 1.1, it was found that people didn't like very old artistic styles, so we avoided using many vintage images for the training set this time.

Added tags to the training set. After researching online, I found two approaches to style transfer: one is to retain only trigger words, while the other is to label all definable elements. This time, I tried the latter approach, and the results were satisfactory but still have room for improvement (to be honest, it was too tedious to manually tag everything).

Enhancements:

Compared to version 1.1, the artistic style is more pronounced, and even using just 1 girl can produce noticeable effects.

Fixed the issue of colors being less vibrant and the overall tone being too dark in version 1.1. Furthermore, enhancements were made to increase color contrast in the visuals of the 2.0 version.

Three resolutions are supported: 512 '512, 512' 768 and 768 '512.

Drawbacks:

The hand parts are prone to breaking. I have tried various methods, but none has yielded satisfactory results. It may help to introduce more negative keywords to reduce the probability of hand parts breaking.

I plan to continue optimizing this model until version 3.0. If you like it, please give me a ❤️. :D