The generator loss is also the same as for the vanilla GAN:
We can compute it by using the following code:
G_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logits_fake, labels=tf.ones_like(D_logits_fake)))
The generator loss is also the same as for the vanilla GAN:
We can compute it by using the following code:
G_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logits_fake, labels=tf.ones_like(D_logits_fake)))