画質 高画質

of some witches in training 🧹 ✨

0 2

YouTube Speedpaint -> https://t.co/Nrltu1xGJL

I had a raffle going on my Discord and this was the winners final piece, their Acrocantho! B)
Sticking to my 2023 goal of training painting more again :3

My Discord server -> https://t.co/aKEd76q9Wg

31 188

some of the decals under training mode directories, these were meant to be used in a plethora of early training maps

3 47

Training sketches to know to draw the Omnic “Moses” and... I had fun giving it more human expressions (the ironyyyyyyyyyyyyyyyyyyyy !🤣)

193 793

Intro: I’ve never seen a form like that it looks strong I better go all out!

Victory: wow….. you were strong….. guess I still have some more training to do I hope we can fight again some day

0 1

Intro: You’re different from other people I’ve seen let’s see what you’ve got!

Victory: that was super fun! Thanks for the training!

0 1

hi! I'm bella pawesborne and im a shiba inu vtuber! I love drawing, singing, and playing video games. I actually have had formal training in classical voice, so I rlly like opera and singing operatically!

0 2

Good Morning, have a lovely day. How’s the training going ?

0 0

im in the art training room, have a wolf

0 10

Goku went from, "There's a point where extra training is just torturing your body," to, "Lets train for 3 years in the room for the heck of it." Lol.

1 9

11/365

This is BunnyMan, a 42 year old deadbeat father who is a coward who uses guns to fight crime. He gets his son (sidekick) every other weekend.

His nemesis is Joseph Keer (Joker), attorney at law. Who helps uphold his ex wife's restraining order against him

0 1

Intro: Not only do you got a gnarly style but you seem pretty strong. Try your best to satisfy me

Win: That was certainly entertaining, make sure to come back to me anytime you’d like I could always use a training partner 😏

0 1

test5 has slightly more training images than 1.3.0, and the output is close to 1.3.0, but degraded.
In 1.3.0, I did a 512x512 cropping step, but in test5 I did not do that, but used aspect ratio bucketing.
Maybe the same procedure with the exact same dataset would reproduce 1.3.0

0 1

Yes...
I am currently testing a WD1.4 based model that dataset with different contents to verify why v1.3.0's outputs are so nicely.
Even if I increase the number of training images, I cannot exceed the miraculously created 1.3.0. I cannot reproduce it... 😭

0 1