search

SSinyu / WGAN-VGG

Pytorch Implementation of Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss
134 stars 29 forks source link

Not well result #3

Open Angus1996 opened 5 years ago

Angus1996 commented 5 years ago

I run this code successfully. But when i was testing，the pred result in saved fig seemed not right. The region that should be black was white???Is there anyone who know what's the probelm?

stefenmax commented 4 years ago

have u met this question like num_samples should be a positive integer value, but got num_samples=0 could u please tell me how to solve it thx

yanglei0941212 commented 3 years ago

I run this code successfully. But when i was testing，the pred result in saved fig seemed not right. The region that should be black was white???Is there anyone who know what's the probelm?

you should check the function save_model in solver.py,this is a pytorch program,so we should save model as 'WGANVGG_xx}iter.pth' not ‘WGANVGG_xx}iter.ckpt’,because ckpt file is for tensorflow

respect3 commented 2 years ago

I run this code successfully. But when i was testing，the pred result in saved fig seemed not right. The region that should be black was white???Is there anyone who know what's the probelm?

Hi,could you please tell me your email address for better communication?And my email is wangxiangcanwww@163.com

JiangGWei commented 2 years ago

I successfully run the code, but the loss function I get from training is negative. And the prediction results are worse than LDCT. Any idea what the problem is?And my email is 779500270@qq.com

chenjiachengzzz commented 2 years ago

I successfully run the code, but the loss function I get from training is negative. And the prediction results are worse than LDCT. Any idea what the problem is?And my email is 779500270@qq.com Have you solved this problem?

WhoYouAre7 commented 1 year ago

I successfully run the code, but the loss function I get from training is negative. And the prediction results are worse than LDCT. Any idea what the problem is?And my email is 779500270@qq.com Have you solved this problem?

Have you solved this problem?

EvelyneCalista commented 2 months ago

I successfully run the code, but the loss function I get from training is negative. And the prediction results are worse than LDCT. Any idea what the problem is?And my email is 779500270@qq.com Have you solved this problem?

Have you solved this problem?

Hi, have you solved the black image issue in inference?