Closed lisiping0817 closed 5 years ago
Hello.
The issue can happen if you have pretty big input images.
We recommend you to use a --chop
argument which automatically slices your input images into smaller patches.
Thank you.
yes! I try this method and it work.
CUDA_VISIBLE_DEVICES=0 python3 main.py --data_test Demo --scale 4 --n_resblocks 32 --n_feats 256 --res_scale 0.1 --pre_train ../experiment/model/EDSR_x4.pt --test_only --save_results i use this command but RuntimeError: CUDA out of memory. Tried to allocate 10.67 GiB