Skip to content
This repository was archived by the owner on Mar 31, 2025. It is now read-only.

Conversation

@sathish-a
Copy link
Contributor

wrt #41 adding super convergence notebook to examples.

@review-notebook-app
Copy link

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.


Powered by ReviewNB

@kihyuks kihyuks marked this pull request as ready for review October 27, 2020 21:30
@@ -0,0 +1,2061 @@
{
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lr=0.1 is used in line 5 and wondering how this value is found.

Did you set this lr from some findings from find_lr call in the previous cell?


Reply via ReviewNB

Copy link
Contributor Author

@sathish-a sathish-a Oct 28, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes from the figure approximately at learning rate 10^-1 the loss started decreasing.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there a systematic way to get lr instead of eyeballing?
If not, would you add a comment how to select lr in practice?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, I'll do it.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hi @sathish-a, just a kind reminder. Please let us know if there is anything that we could help.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@kihyuks thanks for the reminder. I added it.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thanks!

Added information about how to choose a good learning rate.
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants