Simple GAN with TensorFlow

Note

You can follow along with the code in the Jupyter notebook ch-14a_SimpleGAN.

For building the GAN with TensorFlow, we build three networks, two discriminator models, and one generator model with the following steps:

1. Start by adding the hyper-parameters for defining the network:

# graph hyperparameters
g_learning_rate = 0.00001
d_learning_rate = 0.01
n_x = 784  # number of pixels in the MNIST image 

# number of hidden layers for generator and discriminator
g_n_layers = 3
d_n_layers = 1
# neurons in each hidden layer
g_n_neurons = [256, 512, 1024]
d_n_neurons = [256]

# define parameter ditionary
d_params = {}
g_params = {}

activation = tf.nn.leaky_relu
w_initializer = tf.glorot_uniform_initializer
b_initializer = tf.zeros_initializer

2. Next, define the generator network:

z_p = tf.placeholder(dtype=tf.float32, name='z_p', 
        shape=[None, n_z])
layer = z_p

# add generator network weights, biases and layers
with tf.variable_scope('g'):
    for i in range(0, g_n_layers...