Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Document learning rate schedules #2671

Closed
AlexDBlack opened this issue Jan 11, 2017 · 4 comments
Closed

Document learning rate schedules #2671

AlexDBlack opened this issue Jan 11, 2017 · 4 comments

Comments

@AlexDBlack
Copy link
Contributor

Specifically:
(a) what LR schedules are available, and how to use them
https://github.com/deeplearning4j/deeplearning4j/blob/91a481ae8f5bcb4c9ff3463c1bba2df69d7325d2/deeplearning4j-nn/src/main/java/org/deeplearning4j/nn/conf/LearningRatePolicy.java

and,
(b) What the actual mathematical form of these are (and ideally: with example graphs for common configurations)

@tomthetrainer
Copy link

Would the graphs be something we could screen capture from the UI, or something we would have to write some code for.

@AlexDBlack
Copy link
Contributor Author

We could capture this from the UI, though perhaps it might be better if one of us plots them manually (i.e., based on the mathematical formulas in the implementation) with perhaps a few different settings for each for comparison. That might help users understand them better.
I could look at that perhaps over the weekend, if you want.

@AlexDBlack
Copy link
Contributor Author

LR schedules are covered here: https://deeplearning4j.org/quickref#config-schedules

@lock
Copy link

lock bot commented Sep 22, 2018

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.

@lock lock bot locked and limited conversation to collaborators Sep 22, 2018
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

3 participants