Why hardcoded constants?

[constfilin]

On this line :

Conditional-GAN/model_mnist.py

Line 170 in 1b4d7ea

d2 = tf.nn.relu(batch_normal(fully_connect(d1, output_size=7*7*2*64, scope='gen_fully2'), scope='gen_bn2'))

instead of 7*7*2*64 there should be c1*c1*2*self.batch_size.

Otherwise, why does this create variables and then neglects to use them?

[zhangqianhui]

thanks

[constfilin]

JFYI: please see my code based on your conditional GAN.