qwerdbeta

Results 4 issues of qwerdbeta

Does anyone know the format of the data file with all of the paths when training your own images? I can't figure it out. I tried a .flist file tat...

Do any hard coded or configuration values be better optimized for 512 images versus 256? Do the tensors in the model need to be bigger?

Hello, I get good results when training with square mask. But when I use pconv masks images (black and white where white is masked area), all predictions are just white...

Hi, thanks for sharing! Great work! What steps are necessary to make a 512 pixel square image model and train and eval on that? Thanks!