Hello, I use gray image(only one channel) to train EDSR model. I want to reconstruct 80x80 pictures to 320x320.
.sh Order:
python main.py --model EDSR --data_train DIV2K --data_test DIV2K --scale 4 --n_resblocks 32 --n_feats 256 --res_scale 0.1 --patch_size 320 --n_colors 1 --reset --chop --save_results --save_gt --save_models --save test_model --print_every 20 --epochs 200
Error:
RuntimeError: Given groups=1, weight of size 3 3 1 1, expected input[16, 1, 80, 80] to have 3 channels, but got 1 channels instead
I used 24bits image and 8 bits image, which run into the same error.
dongyan007
commented
1 year ago
Hello, I use gray image(only one channel) to train EDSR model. I want to reconstruct 80x80 pictures to 320x320. .sh Order: python main.py --model EDSR --data_train DIV2K --data_test DIV2K --scale 4 --n_resblocks 32 --n_feats 256 --res_scale 0.1 --patch_size 320 --n_colors 1 --reset --chop --save_results --save_gt --save_models --save test_model --print_every 20 --epochs 200
Error: RuntimeError: Given groups=1, weight of size 3 3 1 1, expected input[16, 1, 80, 80] to have 3 channels, but got 1 channels instead I used 24bits image and 8 bits image, which run into the same error.