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
WGan-gp test in the Celeba dataset. #3
Comments
That code looks correct. It's hard to say without seeing the rest of the code, but if you point me to the repo I can try and debug. |
This repo , https://github.com/zhangqianhui/wgan-gp-debug Thank you ! |
The sample of gerneration https://github.com/zhangqianhui/wgan-gp-debug/blob/master/sample/train_04_12601.png in epoch 6, iter= 20401 https://github.com/zhangqianhui/wgan-gp-debug/blob/master/sample/train_06_20401.png |
Could you share the learning curve? (I.e. negative of the critic's loss) |
That doesn't look good. @igul222 did you ever see something like that? Could you share the full code? Best :) |
@martinarjovsky Whose code ? |
Yours! |
Don't know whether it is related but, in my experiments of wgan-gp the loss of G becomes negative, which is different from original wgan in which loss of G is generally positive. Is that normal? |
@igul222 @martinarjovsky hello , have you found the reason about the face generation for lower quality? |
Hi! I haven't looked at the code yet. Can you run ishaan's code (the one on this repo) and see if it gives the same results? |
@martinarjovsky But , his code have not trained in celeba dataset , so Which architecture i need to use ? Is it Ok to use gan_64x64.py and dcgan's architecure? |
That should be fine. |
I test in this project, and it can generate very realistic face image after training in the celeba data-set. |
Here are some differences I found between your implementation and ours which might be responsible:
Hope this helps! |
@igul222 Thanks , I have solved this problem.! |
Cool! What was the issue? |
@martinarjovsky igul222: Gan.py#61: It looks like self.images and self.fake_images both have shape [self.batch_size, 64, 64, self.channel]. In this case, alpha should have shape [self.batch_size, 1, 1, 1], and also reduction_indices on line 67 should be [1,2,3]. |
and my nextbatch() also have som problem. |
And I think layer normalization is very important. Thanks! |
Hi @zhangqianhui Im new to WGAN-GP |
the critic loss should be negative, because the critic loss means the negative of the divergence between the real samples distribution and fake samples distribution. You should read the wgan-gp paper for more details. And g_loss= - c_loss = - D(fake_image) + D(real_image), but the gradient of D(real_images) will not affect g network, so, g_loss= -D(fake_image) |
Doing Classification after using training wgan? |
Actually I've done the multi-task on Discriminator to not only determine the real/fake problem but also classify Identities and other informations from face which has already come up with a good performance on the protocol such as MPIE. So far I'm trying on merging the WGAN to the implementation. Generated Image (20epoch): Number of critic=5 while training, LR=0.0001 |
hi,now i am also trying to train on CelebA (cropped and resized to 64x64) in WGAN-GP mode . I just modify the DATA_DIR in gan_64x64.py. But there was a mistake like this: |
I test the wgan-gp in the celeba dataset.
But the quality of the generative images is worse than the original dcgan.
and i just change the below code in the basic of w-gan using dcgan generator and discirmator.
And the reason?
The text was updated successfully, but these errors were encountered: