512in, 512out - Githubissues

jianzhangcs / panini

Panini-Net: GAN Prior based Degradation-Aware Feature Interpolation for Face Restoration, AAAI 2022 (PyTorch Code)

111 stars 13 forks source link

Open simplew2011 opened 1 year ago

simplew2011 commented 1 year ago

Can we train models with input and output values of 512

simplew2011 commented 1 year ago

how to prepare train dataset?