| Sanqui 2060 ![]() ![]() ![]() ![]() 💛🤍💜🖤🦉 Level: 87 Posts: 1874/2066 EXP: 6305066 For next: 87708 Since: 12-20-09 Pronouns: any ✨ From: Czechia | Estonia Since last post: 6 days Last activity: 1 day |
|
||
|
* flops, wings involved. level 82 exp 5109138 exppct 53.7% numposts 1874 Originally posted by RanAS It's definitely "confusing" it with the underwater sections of Labyrinth, which have similar colors. But in general, a problem of autoencoders is blurry output, because the cheap solution to "get as close to these images as possible" is to simply produce a blurry mess (a "pixel porridge" as my friend called it) that's sort-of inbetween all of them. I'll be exploring some solutions to this problem but most of them are alleviated by using GANs, or generative adversarial networks, which have been oh-so-popular recently (if you've seen this cat does not exist or similar, those have been generated by GANs). They'll be the next thing I'll be learning, probably. ____________________
In our brief lives, • we've managed to meet. • Treasure this gift, • this precious time that we have.
|















