PyTorch-GAN

PyTorch-GAN copied to clipboard

I want to ask what is the function of the pullaway_loss in ebgan.

mo666666

Why there is nn.softmax before nn.crossentroyloss in infogan?

2

In 111 line of the implementaion of infogan, nn.softmax is added to aux_layer. But the returned data of aux_layer is used to compute crossentropy loss in 277 line. Doesn`t nn.crossentropyloss...

Lucksong

fix bugs

chenyang1999

This is a WGAN-gp version of the LOGAN which I used a modified version of it for my work. It was really improving the result of my WGAN-gp, so I...

Baran-phys

Possible error of relativistic gan

2

in https://github.com/eriklindernoren/PyTorch-GAN/blob/master/implementations/relativistic_gan/relativistic_gan.py ``` if opt.rel_avg_gan: g_loss = adversarial_loss(fake_pred - real_pred.mean(0, keepdim=True), valid) else: g_loss = adversarial_loss(fake_pred - real_pred, valid) # Loss measures generator's ability to fool the discriminator g_loss =...

gregjohnso

Problem in ClusterGAN

2

The problem is in clustergan.py: ```python 443: ge_loss = torch.mean(D_gen) + betan * zn_loss + betac * zc_loss 465 d_loss = torch.mean(D_real) - torch.mean(D_gen) + grad_penalty ``` I think it...

zhiqi-li

Is there a certain reason you update the G first?

1

Hi! First, thank you for sharing your code. I appreciate it. And, I have a question about your code. Is there a specific reason why you update the G first...

yoon28

Model Test in DualGAN ([B, 3, 256, 256] => Discriminator => [B, 1, 30, 30] , instead of [B, 1, 1, 1])

3

I have tested the models in DualGAN and got something unexpected. Given input tensor whose shape is [B, 3, 256, 256], I put it into Discriminator in DualGAN and the...

Sean16SYSU

**IMPROVING THE IMPROVED TRAINING OF WASSERSTEIN GANS: A CONSISTENCY TERM AND ITS DUAL EFFECT** [Paper link](http://arxiv.org/abs/1803.01541) I implemented this paper, as it is quoted in WGAN-div.

Sean16SYSU

Unable to create a 128x128 image in WGAN

1

I'm unable to create a 128x128 image. I changed the img_size parameter, can you let me know what changes should I make to the network to incorporate the same?

rohanpandey