当我使用 WGAN 时,有时损失是负数??莫损失代码:
self.g_loss = -tf.reduce_mean(d_logits_fake)
self.d_loss = tf.reduce_mean(d_logits_fake) - tf.reduce_mean(d_logits_real)+GP
当我使用 WGAN 时,有时损失是负数??莫损失代码:
self.g_loss = -tf.reduce_mean(d_logits_fake)
self.d_loss = tf.reduce_mean(d_logits_fake) - tf.reduce_mean(d_logits_real)+GP