Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / majumderb/rezero issues and pull requests

#19 - Bump torch from 1.4.0 to 2.2.0

Pull Request - State: open - Opened by dependabot[bot] 4 months ago
Labels: dependencies

#18 - Add batch_first, dtype, device arguments

Pull Request - State: open - Opened by lericson almost 2 years ago

#17 - Learning rate of the Param resweight

Issue - State: open - Opened by Polarisjame about 2 years ago

#16 - resweight is almost 0

Issue - State: open - Opened by burcehan about 3 years ago - 1 comment

#15 - Is ReZero applicable to fine-tuning?

Issue - State: open - Opened by encounter1997 over 3 years ago

#14 - weight decay for the resweight?

Issue - State: open - Opened by Kyeongpil about 4 years ago - 2 comments

#13 - Can you relaese the code for ResNet-56 in Table2 ?

Issue - State: open - Opened by cuge1995 over 4 years ago

#11 - Sry guys but your paper is not worth more than zero :)

Issue - State: closed - Opened by AmorfEvo over 4 years ago - 1 comment

#10 - can rezero be applied to cnn ?

Issue - State: closed - Opened by carr123 over 4 years ago - 1 comment

#9 - The order of dropout and *resweight

Issue - State: closed - Opened by OneDirection9 over 4 years ago - 3 comments

#8 - when apply rezero to bert or gpt, get NAN gradients

Issue - State: open - Opened by yyht over 4 years ago - 5 comments

#7 - Does it work in not so deep architectures?

Issue - State: closed - Opened by wotulong over 4 years ago - 3 comments

#6 - Relationship between ReZero and Zero gamma trick

Issue - State: closed - Opened by hukkai over 4 years ago - 2 comments

#5 - does rezero work in machine translation tasks?

Issue - State: closed - Opened by zherowolf over 4 years ago - 3 comments

#4 - rezero with norm

Issue - State: closed - Opened by AllenDun over 4 years ago - 1 comment

#3 - I don't see any other application other than NLP?

Issue - State: closed - Opened by nile649 over 4 years ago - 1 comment

#2 - Update README.md : add BiBTex header

Pull Request - State: closed - Opened by mpariente over 4 years ago

#1 - Can the method be applied to CNN?

Issue - State: closed - Opened by JunMa11 over 4 years ago - 1 comment