Regularization for BiGAN, KLpq, GAN and WGAN inferences #817
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@dustinvtran I'm not confident about the change I made in |
|
What you did is correct for both variational and model parameters in |
dustinvtran
reviewed
Jan 7, 2018
| @@ -90,10 +93,14 @@ def build_loss_and_gradients(self, var_list): | |||
| list(range(1, gradients.shape.ndims)))) | |||
| penalty = self.penalty * tf.reduce_mean(tf.square(slopes - 1.0)) | |||
|
|
|||
| reg_terms_d = tf.losses.get_regularization_losses(scope="Disc") | |||
| reg_terms_all = tf.losses.get_regularization_losses() | |||
There was a problem hiding this comment.
Any reason this one and GANInference uses all but BiGAN's uses scope="Gen"?
There was a problem hiding this comment.
I went by how variable lists are created in the respective files. For GAN and WGAN, scope="Gen" is not used for creating var_list so I assumed it wasn't there.
There was a problem hiding this comment.
That makes sense. Merging now. All looks great.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds regularization in the loss functions for BiGAN, KLpq, GAN and WGAN inferences. This makes further progress towards fixing #529.