search

taki0112 / UGATIT

Official Tensorflow implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation (ICLR 2020)
MIT License
6.17k stars 1.05k forks source link

oom question #82

Closed Jiaxin-Wen closed 3 years ago

Jiaxin-Wen commented 4 years ago

I've met the oom question with a TITAN xp 12GB GPU while running the original model(256px , un-light mode)

How much memory would be enough to succefully run the code , like loading the 100-epoch pre-trained model?

kmswlee commented 4 years ago

i did it in 8G memory

sooperset commented 4 years ago

Inference is possible in 12GB, but training is not possible.

swissalps9 commented 3 years ago

Try doing 128px, non-light. Works on 11GB