Generative Adversarial Networks Projects

The generator network

The generator network is a Conditional Generative Adversarial Network (CGAN). The generator network we are going to create is conditioned on the text-conditioning variable. It takes a random noise vector sampled from a latent space and generates an image with a shape of 64x64x3.

Let's start by writing the code for the generator network:

1. Start by creating an input layer to feed the input (the noise variable) to the network:

input_layer2 = Input(shape=(100, ))

2. Next, concatenate the text-conditioning variable with the noise variable along dimension 1:

gen_input = Concatenate(axis=1)([c, input_layer2])

Here, c is the text-conditioning variable. In the previous step, we have written code to generate text-conditioning ...