[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vt/ - Virtual Youtubers

Search:


View post   

>> No.40655779 [View]
File: 582 KB, 1785x1021, Screenshot 2023-01-09 at 18-02-14 TensorBoard.png [View same] [iqdb] [saucenao] [google]
40655779

>>40654107
Yeah, i was trying training with different parameters, because hll2 kinda sucks. I heard running last epochs without --train_text_encoder can improve things, but after doing a lot of comparisons it looks like training more just makes things worse..
I guess it's time to take a break from this model training bullshit. Math is a scam

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