This repository was archived by the owner on Aug 12, 2024. It is now read-only.

Description
Hello, im experiencing wierd artifacts when doing inference on large images.
Image resolution is w1500 x h1125
It would be really nice if I could get the training code so I can just finetune the model. Unfortunately I think its not released yet.
Im calling the code like this:
"--task", "Enhancement",
"--ckpt_path", "maxim_ckpt_Enhancement_FiveK_checkpoint.npz",
"--input_dir", "maxim/images/Enhancement",
"--output_dir", "maxim/images/Results",
"--has_target=False"
I quite often get this wierd column of darker pixels. Anyone has such problems?

sercontent.com/16538903/223081560-d5453a37-6dc4-4c31-80d7-d1fa341b75ad.jpg)
