Hands-On Deep Learning Algorithms with Python

Defining discriminator

We know that the discriminator, , returns the probability; that is, it will tell us the probability of the given image being real. Along with the input image, , it also takes the conditional variable, , as an input. We define the discriminator also as a simple two-layer feed-forward network:

def discriminator(x, c, reuse=False):    with tf.variable_scope('discriminator', reuse=reuse):

Initialize the weights:

 w_init = tf.contrib.layers.xavier_initializer() ...