-
Notifications
You must be signed in to change notification settings - Fork 161
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
Add NovoGrad optimizer #385
Conversation
Add NovoGrad optimizer
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
There are a couple of conflicts, could you sync and merge so we get this submitted? |
Hello! Yes, I will try to fix as soon as possible |
@@ -149,6 +149,7 @@ Gradient Transforms | |||
scale_by_adam | |||
scale_by_belief | |||
scale_by_factored_rms | |||
scale_by_novograd |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you also add novograd
to the Common Optimizers section?
optax/_src/transform.py
Outdated
mu_dtype: optional `dtype` to be used for the first order accumulator; if | ||
`None` then the `dtype is inferred from `params` and `updates`. | ||
Returns: | ||
An (init_fn, update_fn) tuple. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
"A GradientTransformation
object."
Thanks a lot for the contribution! |
@DT6A could you please add a test for the new optimizer? |
@hbq1 is there any reference how the test should be done? |
Hi! We are in the process of writing instructions on this at the moment - in your case it should be enough to just add the new optimizer to the list of optimizers under test in |
@mkunesch Thanks, added it |
Thank you! |
PiperOrigin-RevId: 482336516
Merged! Thanks again for this great contribution 🎉 |
NovoGrad optimizer was presented in this paper: https://arxiv.org/pdf/1905.11286.pdf
It was used to train the ASR model named Jasper: https://arxiv.org/pdf/1904.03288.pdf