search

xinntao / ESRGAN

ECCV18 Workshops - Enhanced SRGAN. Champion PIRM Challenge on Perceptual Super-Resolution. The training codes are in BasicSR.
https://github.com/xinntao/BasicSR
Apache License 2.0
5.91k stars 1.05k forks source link

RuntimeError: CUDA out of memory. #57

Open overthrowrobotics opened 5 years ago

overthrowrobotics commented 5 years ago

Tried to allocate 7.91 GiB (GPU 0; 10.73 GiB total capacity; 2.56 GiB already allocated; 6.63 GiB free; 761.56 MiB cached)

I'm using two 2080Ti with 11GB each and have CUDA_VISIBLE_DEVICES set to 0,1. Can the model be evaluated with both the memory from both GPUs? Input file is 1920x1080. Maybe I just need to split the file into smaller pieces and then recombine. Thoughts?

xinntao commented 5 years ago

Your input is really large (1920x1080). Yes, you need to split the image into smaller pieces and then recombine.