KerasGAN
KerasGAN copied to clipboard
osh
A couple of simple GANs in Keras
I have read your paper recently ,in this paper, you use the variational conditional GAN to approximate robust probability distributions for the task of approximating accurate conditional channel distribution p(y|x)...
Hi, I've got the code running in Python 3.6/Keras 2.0 (Tensorflow backend). So far (epoch ~5000 out of the first 6000) there seems to be no convergence whatsoever, at least...
Hi, I did not figure out why you used the `K.set_image_dim_ordering('th')` It seems you used tensorflow, but why did you set the function to "Theano"?
You define a function `make_trainable()` which sets every layer's `trainable` attribute to either `True` or `False` and call this repeatedly during training. However, setting `keras.layers.Layer.trainable` doesn't have any effect unless...
With Keras1 now being deprecated, a Keras2 version would be greatly appreciated. I tried converting it myself but Keras2 does not support the batchnorm mode=2 option, so it will probably...
I'm a little new to Generative Adversarial Networks and was wondering why the samples with iPython notebook2 are worse compared to the iPython notebook1. Another question I had was, when...
Why is the generative network output shape is (None, 400, 28, 1). Should not this be (None, 28, 28, 1)? I am referring mnist_gan.py
Hello, I just read your tutorial, it is very nice. I have one question: since you've set learning process for GAN `GAN.compile(...)`, why need to set learning process again for...
Hey there, I noticed that my notebook with the latest theano, keras, tensorflow and cuda, CuDNN dies when theano is imported as T for some weird reason. Not sure why,...
