Pitt-Radar icon indicating copy to clipboard operation
Pitt-Radar copied to clipboard

Published 20 hours ago • verified publisher icon yujheli

model train

Open CHB-learner opened this issue 1 year ago • 1 comments

May I ask about your equipment, parameter settings, number of rounds, etc. during training？

ex：train_Hybrid_SR.py When I reproduced it on your dataset, the PSNR could only reach 29。May I ask if there are any differences in preprocessing and training between your two large-scale datasets

CHB-learner avatar Jan 20 '24 07:01 CHB-learner

I guess it caused by the different using log scale and the max-min clipping. I will need to check.

yujheli avatar Mar 22 '24 07:03 yujheli