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

add a decorator that saves the code to file #589

Merged
merged 5 commits into from
Dec 29, 2022

Conversation

ofrimasad
Copy link
Collaborator

No description provided.

@dagshub
Copy link

dagshub bot commented Dec 28, 2022

Copy link
Collaborator

@shaydeci shaydeci left a comment

Choose a reason for hiding this comment

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

2 comments inline.

Apart from that, we might want to reconsider the design of treating the saved_code as a global variable. On one hand, this means that all of the code will be saved (even the one after calling train() which is desired in case, for example when we have some convert script), but there's another concern when launching consecutive trainings (not really a concern, but if we launch Resnet18 recipe then Resnet50 recipe- unnecessary code would be logged). Don't we want just the places we imported (in the sane fashion we do for register?)
Also, would be nice to have a simple unit test, maybe just a sanity that prints some code to see nothing is broken (if it is, it also mean we'd break sg_logger and therefore training).

Copy link
Collaborator

@shaydeci shaydeci left a comment

Choose a reason for hiding this comment

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

LGTM

@shaydeci shaydeci merged commit 8be431d into master Dec 29, 2022
@shaydeci shaydeci deleted the feature/SG-0000_svae_to_file_decorator branch December 29, 2022 11:20
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.

None yet

2 participants