Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

test配置问题 #15

Open
Salmashi opened this issue Jun 2, 2021 · 6 comments
Open

test配置问题 #15

Salmashi opened this issue Jun 2, 2021 · 6 comments

Comments

@Salmashi
Copy link

Salmashi commented Jun 2, 2021

您好 ,请问论文表2 中的kernel width 与test.sh 中 lambda_1,lambda_2的关系是什么呢?

@LongguangWang
Copy link
Member

Hi @Salmashi,Table 2是在isotropic Gaussian kernels上进行的对比,表中kernel width对应tesh.sh中的sig,而lambda_1和lambda_2是用来描述anisotropic Gaussian kernels的两个参数,对应Table 3中不同kernel的“高矮胖瘦”。

@Salmashi
Copy link
Author

image
python test.py --test_only
--dir_data='/home/2K_Resolution'
--data_test='Set14'
--model='blindsr'
--scale='4'
--resume=0
--pre_train='/home/paper3_code/DASR-main/experiment/blindsr_x4_bicubic_aniso/model/DASR_ori.pt'
--blur_type='iso_gaussian'
--sig=1.2
--noise=0.0
--theta=0.0
--lambda_1=0.0
--lambda_2=0.0

[Epoch 0---Set14 x4.0] PSNR: 27.996 SSIM: 0.7631

作者您好,我们加载了您提供的预训练模型(DASR_ori.pt),在如上的配置下跑不出论文表2中的结果,请问为了得到表2中的效果,上述参数应该怎么配置呢?

感谢您上次的耐心解答!

期待您的回复!

@LongguangWang
Copy link
Member

Hi @Salmashi, 抱歉回复晚了。blindsr_x4_bicubic_aniso/model/DASR_ori.pt是在aniso_gaussian模糊核上训练的,请使用我们提供的blindsr_x2_bicubic_iso/model/model_600.pt这个模型来在iso_gaussian模糊核上进行测试以复现表2中的结果。

@Salmashi
Copy link
Author

image
谢谢您的解答,还有一个问题,表上中这个几个核的参数:--theta=0 \ --lambda_1=0.2 --lambda_2=0.2,分别是怎么设置的呢?非常感谢您每一次耐心地解答!

@Lewis0427
Copy link

image
谢谢您的解答,还有一个问题,表上中这个几个核的参数:--theta=0 \ --lambda_1=0.2 --lambda_2=0.2,分别是怎么设置的呢?非常感谢您每一次耐心地解答!

you solve this problem? how to do this

@LongguangWang
Copy link
Member

Hi @Lewis0427 @Salmashi, please refer to this comment for detailed settings of the blur kernels in Tab. 3.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants