Urusan on Nostr: There's really a lot of work that goes into making a truly good dataset. It's kinda ...
There's really a lot of work that goes into making a truly good dataset.
It's kinda funny because the sort of thesis of my main work in this area is that making rough and ready "good enough" stuff is really easy, and that people should just jump into training.
After all, all the images I already posted were derived from a single image (and the base model).
It's adding a more refined touch to it that takes a bunch of time and effort.
It's kinda funny because the sort of thesis of my main work in this area is that making rough and ready "good enough" stuff is really easy, and that people should just jump into training.
After all, all the images I already posted were derived from a single image (and the base model).
It's adding a more refined touch to it that takes a bunch of time and effort.