Skip to content

Conversation

@harsh306
Copy link

#change
return keras.layers.LeakyReLU(alpha=alpha).call(inputs)
#to
return tf.keras.layers.LeakyReLU(alpha=alpha).call(inputs)

Harsh Nilesh Pathak and others added 3 commits June 12, 2018 14:49
#change 
return keras.layers.LeakyReLU(alpha=alpha).call(inputs)
#to
return tf.keras.layers.LeakyReLU(alpha=alpha).call(inputs)
@harsh306 harsh306 changed the title [RESOLVED]: name 'keras' is not defined. [RESOLVED] errors and evaluation metric added Jun 21, 2018
harsh306 and others added 9 commits June 22, 2018 15:41
I am sure about SRResNet thanks to 
KJLiu@0bb3347
I have extended this to SRGAN function but I am not very sure. Because in GAN, discriminator and generator train in turns and for distribution amongst tower generator should wait for discriminator. I have used optimizer in discriminator 2 times once inside the tower and once outside with average gradients.   Would be glad to get some open source help or comments  if not correct.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant