PyTorch-GAN
PyTorch-GAN copied to clipboard
PyTorch implementations of Generative Adversarial Networks.
I'm getting the error: `ValueError: Target and input must have the same number of elements. target nelement (64) != input nelement (32)`, traceable back to line 470 in clustergan.py. I...
Pull request for issue #92
Hi, I think the `loss_GAN` in [`esrgan.py`](https://github.com/eriklindernoren/PyTorch-GAN/blob/a163b82beff3d01688d8315a3fd39080400e7c01/implementations/esrgan/esrgan.py#L135) is missing a term. In the [ESRGAN paper](https://arxiv.org/abs/1809.00219) the adversarial loss is like the discriminator loss a sum of two terms.  However...
I have been working on this GAN can please , give instructions to use this GAN for my own inage dataset , when I used it on my own images...
DCGAN fails learning the mnist dataset. Is there a problem in implementation.
Hi! I just implemented A GAN based missing value imputation framwork, please see relative paper in readme :)
hey! The paper mentions using 0.5 dropout on all but the last layer but this implementation uses 0.25 on all layers of the discriminator. Also, strided convolutions every odd layer....
Hi, I am a newbie and I am trying to train using the MRI2D dataset. My data set has saved 2D slices in a matrix form with the shape [500,...
parse_args throws an error
The object detection training set is too small. How can I use GAN to generate the object detection training set?