search

sanghyun-son / EDSR-PyTorch

PyTorch version of the paper 'Enhanced Deep Residual Networks for Single Image Super-Resolution' (CVPRW 2017)
MIT License
2.43k stars 669 forks source link

the result on BSD100 #72

Open liwanlan opened 5 years ago

liwanlan commented 5 years ago

Hi~ I want to know why the PSNR and SSIM values on BSD100 when I run the pre-trained code are worse than your published a lot . Thank you.

sanghyun-son commented 5 years ago

Hello.

I checked the code and results were consistent with our paper.

Please note that our paper report the results from large EDSR model (32 ResBlocks + 256 channels), not smaller one (16 ResBlocks + 64 channels).

Thank you.

wbhu commented 5 years ago

Hi @thstkdgus35,

If I put the 'benchmark' folder in src/data/, and I want to generate the benchmark results reported in the paper, I run 'python main.py --data_test Set5 --scale 4 --n_resblocks 32 --n_feats 256 --res_scale 0.1 --pre_train ../experiment/model/EDSR_x4.pt --test_only --self_ensemble --dir_data data' in src folder, but the results show that none images were tested. What should I do?

Thx a lot!

taniaCHY commented 5 years ago

Can you send me a copy of your DIV2K test results? Email 1964771130 @qq.com

liu1216 commented 2 years ago

嗨@thstkdgus35，

如果我将 'benchmark' 文件夹放在 src/data/ 中，并且我想生成论文中报告的基准测试结果，我运行 'python main.py --data_test Set5 --scale 4 --n_resblocks 32 --n_feats 256 --res_scale 0.1 --pre_train ../experiment/model/EDSR_x4.pt --test_only --self_ensemble --dir_data data' 在 src 文件夹中，但结果显示没有图像经过测试。我应该怎么办？

多谢！

Excuse me, have you solved it

taniaCHY commented 2 years ago

这是来自QQ邮箱的自动回复邮件。   你好，你的邮件已收到！