[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]

/vt/ - Virtual Youtubers

Search:


View post   

>> No.67979898 [View]
File: 3.70 MB, 1496x2184, 1702725600804692.png [View same] [iqdb] [saucenao] [google]
67979898

>>67969097
Sorry for the late reply, fell asleep kek
I am using the colab this anon posted seems didn't change any of his settings and it seems to be working great
>>67620320

For the dataset, I am doing the same thing I was for 1.5 tagging with this and making sure to remove color facial feature tags (for FuwaMoco I did them in one LoRA so went a bit further and removed all facial feature tags and let it burn into their name).
For styles same thing except added score_9 and source_anime in front of all the images.
https://github.com/picobyte/stable-diffusion-webui-wd14-tagger

>>67971504
>>67972066
Cute Anyas
>>67974926
Oh shit thanks anon, will test it out!
>>67976868
>Hm? I'm still here kek.
A very likely story....
>>67978357
Taki was interesting experience for childhood me kek
>>67978465
G'morning! Cute!

Navigation
View posts[+24][+48][+96]