generative-adversarial-network

Here is my discriminator architecture: def build_discriminator(img_shape,embedding_shape): model1 = Sequential() model1.add(Conv2D(32, kernel_size=5, strides=2, input_shape=img_shape, padding="same")) model1.add(LeakyReLU(alpha=0.2)) model1.add(Dropout(0.25)) model1.add(Conv2D(48, kernel_size=5, strides=2, padding="same")) #model.add(ZeroPadding2D(padding=((0,1),(0,1)))) model1.add(BatchNormalization(momentum=0.8)) model1.add(LeakyReLU(alpha=0.2)) model1.add(Dropout(0.25)) model1.add(Conv2D(64, kernel_size=5, strides=2, padding="same")) model1.add(BatchNormalization(momentum=0.8)) model1.add

I came across a strange issue when using keras to implement GAN model. With GAN we need to build up G and D first, and then add a new Sequential model (GAN) and add(G), add(D) sequentially afterwards. Keras seems to backprop back to G (via GAN model) when I do D.train_on_batch , and I got an InvalidArgumentError: You must feed a value for placeholder tensor 'dense_input_1' with dtype float . If I remove the GAN model (the last stacked G then D sequential model), it computes d_loss correctly. My environment is: Ubuntu 16.04 keras 1.2.2 tensorflow-gpu 1.0.0 keras config: { "backend": "tensorflow

I am reading people's implementation of DCGAN, especially this one in tensorflow. In that implementation, the author draws the losses of the discriminator and of the generator, which is shown below (images come from https://github.com/carpedm20/DCGAN-tensorflow ): Both the losses of the discriminator and of the generator don't seem to follow any pattern. Unlike general neural networks, whose loss decreases along with the increase of training iteration. How to interpret the loss when training GANs? Unfortunately, like you've said for GANs the losses are very non-intuitive. Mostly it happens