Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

【使用反馈】模型识别率太感人了 #15

Open
Xian0Grem opened this issue Nov 12, 2024 · 5 comments
Open

【使用反馈】模型识别率太感人了 #15

Xian0Grem opened this issue Nov 12, 2024 · 5 comments

Comments

@Xian0Grem
Copy link

Screenshot_20241112_113747
这是默认模型

@Xian0Grem
Copy link
Author

Screenshot_20241112_113608
这是r15那个模型

@iyume
Copy link
Owner

iyume commented Nov 12, 2024

你可以在这里打个断点或者 print 看看:

nsfw-model 的判定条件是五个条件中最高概率在 ["hentai", "porn", "sexy"] 之中就是 positive。

你可以自己加点判断条件,这个就比较困难了。而且只拿反例来讲没什么意义,得统计一下错误率,3-5% 我觉得都是良好的。

至于 safety-checker 那边也不方便调整。

@Xian0Grem
Copy link
Author

Xian0Grem commented Nov 12, 2024

哦,抱歉我举的例子不太全面
非常感谢你提供方法,虽然我完全不会用)
我存储的图每天都清理,我这里只能提供今天的数据了
网页捕获_12-11-2024_143457_wasesuki top
Q版表情包的占比特别高
因为我bot接收到的信息色图少,q版表情多
而这个模型会把部分q版表情包识别成色图
目前我只是把本插件作为静默存色图用
如果投入管理功能)有点吓人

@iyume
Copy link
Owner

iyume commented Nov 15, 2024

近期有空会添加 https://huggingface.co/Falconsai/nsfw_image_detection

@iyume
Copy link
Owner

iyume commented Nov 18, 2024

0.13 已发版,新增 nsfw_image_detection 模型。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants