New answers tagged generative-adversarial-networks
1
vote
Accepted
"Window Pane" pattern when training CycleGAN
I found the problem:
After converting the images to tensors I was using torch.reshape to try swapping the order of <width, height, channels> to <channels, width, height>. However, this ...
Top 50 recent answers are included
Related Tags
generative-adversarial-networks × 263deep-learning × 68
neural-networks × 47
generative-model × 47
machine-learning × 45
convolutional-neural-networks × 30
image-generation × 22
papers × 19
objective-functions × 19
training × 16
wasserstein-gan × 16
computer-vision × 14
variational-autoencoder × 13
discriminator × 12
terminology × 11
pytorch × 9
generator × 9
math × 8
applications × 8
image-processing × 8
datasets × 7
autoencoders × 7
loss × 7
dc-gan × 7
natural-language-processing × 6