Open wangguanan opened 7 years ago
The CIFAR10 model which got the best results was a resnet, but not 101-resnet. We trained 101-resnet on bedrooms (see gan_64x64.py) to demonstrate that it worked, but we never expected that model to produce particularly great samples -- it's probably too deep and has too many bottleneck layers for that.
From: wangguanan notifications@github.com Sent: Friday, August 11, 2017 11:26:32 PM To: igul222/improved_wgan_training Cc: Subscribed Subject: [igul222/improved_wgan_training] how to adapt gan_cifar_resnet.py to Resnet101 Implementation (#42)
Thanks for the author's wonderful theory and model, and I have 3 questions about Improved WGAN model with resnet 101 discriminator: Firstly, according to my understanding, your implementation of WGAN model in file adapt gan_cifar_resnet.py took of actually discriminator resnet of 14 layers( 4 residual blocks 3 = 12 layers, input layers and output layers), generator renset of 15 layers(3 residual blocks 3 = 9 layers, input layer , one conv layer and output layer). Is there any error about on my understanding? [image]https://user-images.githubusercontent.com/18496599/29239527-18e371c4-7f83-11e7-91e2-2e69fb27b53b.png
[image]https://user-images.githubusercontent.com/18496599/29239461-c6c60420-7f81-11e7-9f4b-f78fe565ad1a.png
Secondly, if my understanding is right, how can i modify the model to one with resnet101 discriminator?
Thanks !
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/igul222/improved_wgan_training/issues/42, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ABBP7gmZ-xIVGV5aomJjsnloTVmTixTwks5sXW_IgaJpZM4O1XM5.
Thanks!
Thanks for the author's wonderful theory analysis and beautiful model implementation. I have 3 questions about Improved WGAN model with resnet 101 discriminator.
Firstly, according to my understanding, implementation of improved wgan model in file
adapt gan_cifar_resnet.py
took of actually discriminator resnet of 14 layers( 4 residual blocks 3 = 12 layers, input layers and output layers), generator renset of 15 layers(3 residual blocks 3 = 9 layers, input layer , one conv layer and output layer). Is there any error about on my understanding?Secondly, if my understanding is right, how can i modify the model to one with resnet101 discriminator?
Thirdly, is there any difference between implementation of your resnet and
tensorflow.contrib.slim.nets.resnet_v1
? Can i replace your resnet withtensorflow.contrib.slim.nets.resnet_v1
or others?Thanks !