JingyunLiang / SwinIR

SwinIR: Image Restoration Using Swin Transformer (official repository)

Home Page:https://arxiv.org/abs/2108.10257

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Training Issues

msx-123 opened this issue · comments

First of all, thank you for your excellent work.
Currently I have two questions.

  1. The training demo you gave has the training in the main_train_psnr.py file, but I don't find this main_train_psnr file in the project, how do I train it?
  2. I am more concerned about the training configuration and training time, how long did you train the DIV2K dataset with classic super resolution using what configuration of graphics card?
    Looking forward to your reply, thanks!

If you meet I/O bottleneck, preparing the dataset as lmdb (https://github.com/cszn/KAIR/blob/master/scripts/data_preparation/create_lmdb.py) or extracting them as small patches (https://github.com/cszn/KAIR/blob/master/scripts/data_preparation/extract_subimages.py) can help a lot. The PSNR performance will the same.

ezoic increase your site revenue