[sched] polynomial_decay_schedule use default power=1.0 #6473
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
sgugger
approved these changes
Aug 17, 2020
Zigur
pushed a commit
to Zigur/transformers
that referenced
this pull request
Oct 26, 2020
fabiocapsouza
pushed a commit
to fabiocapsouza/transformers
that referenced
this pull request
Nov 15, 2020
fabiocapsouza
added a commit
to fabiocapsouza/transformers
that referenced
this pull request
Nov 15, 2020
…ingface#6473)" This reverts commit 636da6c.
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Outdated suggestions cannot be applied.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.
As discussed in #6361 we weren't sure why fairseq's
polynomial_decay_schedulepowerdefault was1.0, and decided to go with2.0as the latter does something polynomial.I got the devs at fairseq to answer this question: facebookresearch/fairseq#2466 (comment)
So, perhaps we do the same or we don't.
If we don't - then the doc needs to be fixed that the default is
power=2.0as currently it says1.0- my mistake. If we do (this PR), then the doc is already correct.Thanks.