-
Notifications
You must be signed in to change notification settings - Fork 1.3k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Improve performance by using a custom causal convolution
tf.atrous_conv2d pads the height dimension of the input to be a multiple of the dilation rate, which explains why the model was slow and used a lot of memory. After switching to the custom causal convolution function in this commit, performance has increased massively, and it's now possible to train large models with dilation rates that wrap around multiple times from 1 to 512.
- Loading branch information
Showing
2 changed files
with
41 additions
and
24 deletions.
There are no files selected for viewing
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
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