Add support for scheduled weight decays in RectifiedAdam. #1974
Conversation
This file contains hidden or 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
RAdam implements weight decay based on AdamW, and the latter supports scheduling for both learning rate and weight decays as part of its warm restarts version. This patch extends existing support of Keras schedulers for the learning rate to weight decay, matching the weight decay features of AdamW.
|
You are owner of some files modified in this pull request. |
This problem appears to be already affecting the deserialization of learning rate schedulers regardless of this patch.
WindQAQ
reviewed
Jul 7, 2020
ashutosh1919
pushed a commit
to ashutosh1919/addons
that referenced
this pull request
Jul 12, 2020
…#1974) * Add support for scheduled weight decays in RectifiedAdam. RAdam implements weight decay based on AdamW, and the latter supports scheduling for both learning rate and weight decays as part of its warm restarts version. This patch extends existing support of Keras schedulers for the learning rate to weight decay, matching the weight decay features of AdamW. * Fix code style issues. * Fix deserialization of schedulers. This problem appears to be already affecting the deserialization of learning rate schedulers regardless of this patch. * Fix comparison when using the optimizer inside a tf.function.
jrruijli
pushed a commit
to jrruijli/addons
that referenced
this pull request
Dec 23, 2020
…#1974) * Add support for scheduled weight decays in RectifiedAdam. RAdam implements weight decay based on AdamW, and the latter supports scheduling for both learning rate and weight decays as part of its warm restarts version. This patch extends existing support of Keras schedulers for the learning rate to weight decay, matching the weight decay features of AdamW. * Fix code style issues. * Fix deserialization of schedulers. This problem appears to be already affecting the deserialization of learning rate schedulers regardless of this patch. * Fix comparison when using the optimizer inside a tf.function.
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.
RAdam implements weight decay based on AdamW, and the latter supports scheduling for both learning rate and weight decays.
This patch extends existing support of Keras schedulers for the learning rate to weight decay, matching the weight decay features of AdamW.
fixes #1908