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

Follow PyTorch Guide of updating Learning Rate #1101

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

hyecheol123
Copy link

#1100

I modified the code slightly, adding conditional statement to avoid calling decaying function on the first epoch

@junyanz
Copy link
Owner

junyanz commented Jul 21, 2020

Thanks for the PR. If this condition is added, it seems that the learning rate will not change during the first epoch of contine_train. I am wondering if there is a workaround solution. @hyecheol123 @taesungp

@hyecheol123
Copy link
Author

I did not considered continue training while writing that code and will take a look whether there is a good solution to handle it.

@hyecheol123
Copy link
Author

I have one question regarding the starting epoch without continue_train option.

When continue_train does not specified, does it always starts train from epoch 1?
In this case, discussion from the PyTorch forum might be useful.

@junyanz
Copy link
Owner

junyanz commented Jul 22, 2020

The epoch_count is 1 by default. Without continue_train, it will start as 1.

@hyecheol123
Copy link
Author

I will try to fix the code so that we can get proper learning rate decay for continued train.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants