Fix rolling avg #537
Merged
Fix rolling avg #537
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
|
It is currently not trivial to judge what's going on by looking at the warn() message. Can you elaborate more? |
|
I'm honestly not quite sure ;-) I doubled-checked the math and can't find a mistake there, so numerical problems is the only explanation I have right now. It did happen several times, and this fix helped to prevent the crash and led to reasonably converged models. |
|
I've removed the warning which was mainly for debugging this and instead added a comment explaining what's going on. I'll merge now. |
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Outdated suggestions cannot be applied.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.
The rolling average is used to skip updates for noisy gradients but has caused occasional problems, maybe because of numerical stability issues. This PR implements a workaround.